WO2023219369A1

WO2023219369A1 - Method and system for predicting change in skin brightness for prescription containing hydroquinone

Info

Publication number: WO2023219369A1
Application number: PCT/KR2023/006235
Authority: WO
Inventors: 홍성헌; 송해중; 임상섭; 최선국; 김진현; 이운영
Original assignee: 주식회사 엘지생활건강
Priority date: 2022-05-10
Filing date: 2023-05-09
Publication date: 2023-11-16
Also published as: KR20230157785A

Abstract

The present invention pertains to a method and system for predicting changes in skin brightness for a prescription containing hydroquinone. The method and system for predicting changes in skin brightness for a prescription containing hydroquinone is designed to involve querying single nucleotide polymorphism information of a prediction target object to a skin brightness change prediction model pre-stored on a server to compute predicted skin brightness change information and outputting the computed data. The skin brightness change prediction model is trained to learn first skin brightness changes composed of sets of paired information of SNP information and skin brightness change information when a substance containing hydroquinone is applied to objects with the SNP information. When queried with SNP information, the model outputs predicted skin brightness changes for an object with the given SNP information upon application of a hydroquinone-containing substance.

Description

Method and system for predicting changes in skin brightness of prescriptions containing hydroquinone

The present invention relates to a method and system for predicting changes in skin brightness of a prescription containing hydroquinone.

Hydroquinone is an ingredient used as a skin whitening agent, and is known to suppress excessive pigmentation of the skin, such as spots and freckles. In particular, it is known to decolorize excessively deposited pigment on the skin by reducing the production of melanin pigment in the epidermal layer of the skin.

In general, when a substance containing hydroquinone is applied to the skin, it can be effective in terms of whitening, but the efficacy varies depending on the user, and even if it shows the same efficacy, the degree of irritation to the user is different (Korean Published Patent No. No. 10-2016-0014271).

Meanwhile, it has been revealed that even with the same substance, the change in skin brightness and level of irritation to the user are related to genetic characteristics.

However, there is a lack of research on what genetic factors affect the changes in skin brightness and stimulation level that occur when substances containing hydroquinone are applied.

Under this background, the present inventors used statistical analysis methods and machine learning techniques to select specific single nucleotide polymorphism (SNP) markers that have a significant correlation with the change in skin brightness and irritation level that occurs when a substance containing hydroquinone is applied, , we have completed the present invention that can predict skin brightness, predict stimulation information, and even recommend cosmetics when the user's genetic information is input.

One object of the present invention is to provide a single nucleotide polymorphism (SNP) marker for determining changes in skin brightness with hydroquinone.

One object of the present invention is to provide a single nucleotide polymorphism (SNP) marker for determining hydroquinone skin irritation.

Another object of the present invention is to provide a composition for determining changes in hydroquinone skin brightness, which includes a probe capable of detecting or an agent capable of amplifying a single nucleotide polymorphism (SNP) marker for determining changes in hydroquinone skin brightness.

Another object of the present invention is to provide a composition for determining hydroquinone skin irritation, which includes a probe capable of detecting or an agent capable of amplifying a single nucleotide polymorphism (SNP) marker for determining hydroquinone skin irritation.

Another object of the present invention is to provide a kit or microarray for determining changes in hydroquinone skin brightness, including the composition.

Another object of the present invention is to provide a kit or microarray for determining hydroquinone skin irritation containing the composition.

Another object of the present invention is to provide a composition for anti-irritation by hydroquinone containing an irritation reliever as an active ingredient.

Another object of the present invention is to provide a cosmetic composition, quasi-drug composition, or pharmaceutical composition containing the composition as an active ingredient.

Another object of the present invention is to provide a method of providing information for selecting a customized irritation reliever for alleviating irritation caused by hydroquinone, which includes the step of identifying the polymorphic site of the single nucleotide polymorphism marker for determining hydroquinone skin irritation.

Another object of the present invention is to provide a method of providing information on changes in skin brightness due to hydroquinone, including the step of identifying the polymorphic site of the single nucleotide polymorphism marker.

Another object of the present invention is to provide a method of providing information on hydroquinone skin irritation, including the step of identifying the polymorphic site of the single nucleotide polymorphism marker.

Another object of the present invention is to provide a method for predicting changes in skin brightness of a prescription containing hydroquinone.

Another object of the present invention is to provide a method for predicting the presence or absence of irritation of a prescription containing hydroquinone.

Another object of the present invention is to provide a method for predicting the critical irritation score of a prescription containing hydroquinone.

Another object of the present invention is to provide a method for selecting a prescription based on the predicted change in skin brightness of a prescription containing hydroquinone.

Another object of the present invention is to provide a method for selecting a prescription based on predictive stimulation information of a prescription containing hydroquinone.

Another object of the present invention is to provide a method for predicting the presence or absence of irritation based on the critical irritation score of a prescription containing hydroquinone.

Another object of the present invention is to provide single nucleotide polymorphism (SNPs) markers that affect changes in skin brightness when a substance containing hydroquinone is applied according to a prescription.

Another object of the present invention is to provide single nucleotide polymorphism (SNPs) markers that affect skin irritation when applying a substance containing hydroquinone according to a prescription.

Each description and embodiment disclosed in the present invention can also be applied to each other description and embodiment. That is, all combinations of the various elements disclosed in the present invention fall within the scope of the present invention. Additionally, the scope of the present invention cannot be considered limited by the specific description described below.

One aspect of the present invention provides a single nucleotide polymorphism (SNP) marker for determining changes in hydroquinone skin brightness.

One aspect of the present invention provides a single nucleotide polymorphism (SNP) marker for determining hydroquinone skin irritation.

Another aspect of the present invention provides a composition for determining changes in hydroquinone skin brightness, comprising a probe capable of detecting or an agent capable of amplifying a single nucleotide polymorphism (SNP) marker for determining changes in hydroquinone skin brightness.

Another aspect of the present invention provides a composition for determining hydroquinone stimulation, comprising a probe capable of detecting or an agent capable of amplifying a single nucleotide polymorphism (SNP) marker for determining hydroquinone stimulation.

In the present invention, the term "polymorphism" refers to the presence of two or more alleles at one genetic locus. Among the polymorphic sites, only a single base differs from person to person, and is referred to as a single nucleotide polymorphism. It is called a (single nucleotide polymorphism, SNP). Preferred polymorphic markers have two or more alleles that exhibit an occurrence frequency of more than 1%, more specifically, more than 10% or 20% in the selected population. A 'genetic polymorphism marker' is generally one of two or more alleles at the same genetic position (base). This refers to a case where an allele is observed. Generally, depending on the individual, there is a major allele/major allele, a major allele/minor allele, or a minor allele. There is a case of minor allele/minor allele. In the present invention, it may be used interchangeably with “polymorphic marker” and may mean the base and base region of a lower allele, or may be defined together with the number and base position of the chromosome, but is not limited thereto.

In the present invention, the term “allele” refers to several types of one gene that exist at the same genetic locus on a homologous chromosome. Alleles are also used to represent polymorphisms, for example, a SNP has two types of alleles (bialleles). In addition, it refers to a combination of two or more bases with the same number and base position of the chromosome, and the bases are a major allele with a high frequency of occurrence in a specific group of individuals and a major allele with a lower frequency of occurrence than the major allele. Contains alleles (minor alleles).

Specifically, the genetic polymorphism markers of the present invention are significantly associated with hydroquinone skin brightness changes and/or irritation, and when one or more of the two alleles (alleles) have one or more minor alleles, hydroquinone skin It can be said that brightness changes and/or stimulation are more significant compared to individuals with the major allele. That is, in the case of major allele/minor allele, minor allele/minor allele, major allele/major allele ), it can be seen that the degree of skin brightness change due to hydroquinone is high or has low skin characteristics, and the degree of stimulation by hydroquinone is high or low.

The single nucleotide polymorphism marker of the present invention allows predicting an individual's unique hydroquinone skin brightness change and/or irritation characteristics, so it can also provide information on active ingredients that effectively affect skin brightness change/skin irritation change caused by hydroquinone. As such, it may be possible to provide personalized skin care product information and lifestyle pattern information, but is not limited to this.

In the present invention, the term "rs_id" refers to rs-ID, an independent marker assigned to all SNPs initially registered by NCBI, which began accumulating SNP information in 1998. rs_id described in this table refers to a SNP marker, which is a polymorphic marker of the present invention.

The single nucleotide polymorphism marker may be one or more single nucleotide polymorphism markers selected from the single nucleotide polymorphism markers shown in Tables 1 to 3. The single nucleotide polymorphism marker shown in Table 1 may be used to determine the degree of correlation with the degree of hydroquinone skin brightness. The single nucleotide polymorphism marker shown in Table 2 may be used to determine the degree of correlation with the degree of hydroquinone stimulation. The single nucleotide polymorphism marker shown in Table 3 may be used to determine the degree of correlation with the presence or absence of hydroquinone stimulation.

Specifically, the single nucleotide polymorphism marker may be one or more single nucleotide polymorphism markers selected from Tables 1 to 3 related to the degree of skin brightness by hydroquinone/degree of stimulation by hydroquinone/existence or absence of stimulation by hydroquinone.

The significance of the single nucleotide polymorphism marker of the present invention for skin brightness change/irritation due to hydroquinone is less than 0.1, less than 0.05, less than 0.01, less than 0.001, less than 0.0001, less than 0.00001, less than 0.000001, less than 0.0000001, less than 0.00000001, or 0.0000000. less than 01 Characterized by, but not limited to, p-values such as p-value. Specifically, the p-value may be less than 0.1, and more specifically, the p-value may be less than 0.01, but is not limited thereto.

The single nucleotide polymorphism (SNP) marker of the present invention may be any one or more selected from the markers shown in Tables 1 to 3, but is not limited thereto. The single nucleotide polymorphism (SNP) marker may be one or more, and may be used in combinations such as two or more, three or more, four or more to determine the degree of hydroquinone skin brightness change/stimulation, but is limited thereto. It doesn't work.

The marker may be, but is not limited to, the SNP itself, a polynucleotide consisting of 5-100 contiguous DNA sequences containing the SNP position, or a polynucleotide consisting of a complementary sequence thereof.

In one embodiment, the single nucleotide polymorphism marker may be one or more selected from the markers shown in Table 1 related to changes in skin brightness due to hydroquinone, but is not limited thereto.

A description of the marker selected among the markers shown in Table 1 may be as follows.

For example, if the SNP ID is rs7718428, Chr.Position (GRCh ver. 37) is described as "5:149167073", and Allele is described as A>G, this is base 149167073 of human chromosome 5. indicates that it is A or G, and the base located to the left of the ">" of the allele may indicate the higher allele (major allele), and the base located to the right may indicate the lower allele (minor allele).

In one embodiment, the marker selected in Table 1 is a polynucleotide consisting of 5-100 consecutive DNA sequences containing the 110257814th base of human chromosome 1, where the 110257814th base is C or G (rs78955679). nucleotide; A polynucleotide consisting of 5-100 consecutive DNA sequences containing the 69913239th base of human chromosome 3, where the 69913239th base is T or C (rs73838671); A polynucleotide consisting of 5-100 consecutive DNA sequences containing base 140601982 of human chromosome 4, where base 140601982 is A or G (rs73854327); A polynucleotide consisting of 5-100 contiguous DNA sequences containing base 149167073 of human chromosome 5, where base 149167073 is A or G (rs7718428); A polynucleotide consisting of 5-100 contiguous DNA sequences containing the 52666448th base of human chromosome 6, where the 52666448th base is C or G (rs11970311); A polynucleotide consisting of 5-100 consecutive DNA sequences containing base 88976963 of human chromosome 11, where base 88976963 is T or C (rs12788260); A polynucleotide consisting of 5-100 consecutive DNA sequences containing the 16519020th base of human chromosome 12, where the 16519020th base is A or G (rs2160512); A polynucleotide consisting of 5-100 contiguous DNA sequences containing the 28304344th base of human chromosome 15, where the 28304344th base is A or G (rs7162117); and one or more polynucleotides selected from the group consisting of complementary polynucleotides thereof, but is not limited thereto. The markers described above are only a partial example in Table 1, and can be selected in the same manner as above on chromosomes at other locations.

In another embodiment, as in Table 1, any one or more of the single nucleotide polymorphism (SNP) markers shown in Table 2 may be selected, but is not limited thereto.

In another embodiment, as in Table 1, any one or more of the single nucleotide polymorphism (SNP) markers shown in Table 3 may be selected, but is not limited thereto.

The allele of the present invention has the same number of chromosomes in each individual, and among them, there is a major allele and a minor allele of the SNP, and the bases of the polymorphic site of the polymorphic marker are in the lower allele. As the alleles increase one by one, the higher alleles can decrease by one, and as the higher alleles increase one by one, the lower alleles can decrease by one. However, the range in which the minor allele and major allele can increase and decrease is i) major allele/major allele, ii) major allele/minor allele. ), iii) It can be within three types of minor allele/minor allele, and the allele may decrease or increase within the range of the three types, but is not limited thereto. .

In addition, in the present invention, the marker is a marker that can determine changes in skin brightness and/or irritation caused by hydroquinone as the bases of the polymorphic site of the polymorphic marker of an individual increase by one minor allele. Specifically, among the two alleles (allele), there is at least one minor allele (1) major allele/minor allele or (2) minor allele ( Individuals with a minor allele have a lower degree of change in skin brightness caused by hydroquinone compared to those with a major allele, which is a typical individual. It can be judged to have high or low skin characteristics, or the degree of irritation by hydroquinone can be judged to have high or low skin characteristics.

More specifically, as the number of minor alleles among the markers shown in Table 1 increases one by one, the degree of change in skin brightness caused by hydroquinone can be determined. As an example, if among the markers shown in Table 1, the upper allele is A at base 149167073 of the individual's chromosome 5, and the lower allele is G (rs7718428), compared to a person possessing A/A, A/ In the case of having G or G/G, the effect size is negative (-), so it can be judged that the effect of improving skin brightness by hydroquinone is reduced, and the upper allele at base 88976963 of the individual's chromosome 11 is T. , If the sub-allele is C (rs12788260), compared to a person with TT, if you have T/C or C/C, the effect size is positive (+), so the effect of improving skin brightness by hydroquinone is higher. judged to be increasing may, but is not limited to this.

Among the markers shown in Table 2, as the minor allele increases one by one, the degree of increase or decrease in the degree of hydroquinone stimulation can be determined. For example, among the markers shown in Table 2, if the upper allele is T at base 46332169 of chromosome 2 of an individual and the lower allele is C (rs12712969), compared to a person possessing T/T, T/ In the case of having C or C/C, the effect size is positive, so it can be judged that the degree of stimulation by hydroquinone increases, and the upper allele is T at the 46024212th base of the individual's chromosome 2, and the lower allele is When the gene is C (rs6720975), compared to people who have T/T, the effect size is negative (-) in those who have T/C or C/C, so the degree of stimulation by hydroquinone is believed to be reduced. You may judge, but you are not limited to this.

Among the markers shown in Table 3, the probability of stimulation by hydroquinone can be determined as the number of minor alleles increases. As an example, among the markers shown in Table 3, if the upper allele is A at base 46018223 of the individual's chromosome 2 and the lower allele is C (rs11898074), compared to a person possessing A/A, A/ If you have C or C/C, the effect size is positive, so it can be judged that the probability of stimulation by hydroquinone is high, and the upper allele is A at base 46160405 of the individual's chromosome 2, and the lower allele is When is G (rs12999695), compared to people who have A/A, the effect size is negative (-) in those who have A/G or G/G, so it can be judged that the probability of stimulation by hydroquinone is low. may, but is not limited to this.

The above bases are listed as an example of one polymorphic marker in Tables 1 to 3, and are not described in detail, but can be interpreted and derived as described above.

In the present invention, the term "probe capable of detecting a marker for determining hydroquinone skin brightness change/irritation" refers to the degree of skin brightness change/skin irritation caused by hydroquinone by specifically identifying the polymorphic region of the above gene through a hybridization reaction. refers to a composition capable of diagnosing, and the specific method of such genetic analysis is not particularly limited and may be any gene detection method known in the technical field to which this invention pertains. Additionally, the above terms can be used interchangeably with the terms ‘for determining changes in skin brightness due to hydroquinone/for determining the degree of skin irritation due to hydroquinone’.

In the present invention, the term "agent capable of amplifying a marker for determining hydroquinone skin brightness change/irritation" means that the degree of skin brightness change/skin irritation due to hydroquinone can be diagnosed by confirming the polymorphic site of the above gene through amplification. It refers to a composition that has a composition, and specifically refers to a primer that can specifically amplify the polynucleotide of the marker for determining the hydroquinone skin brightness change/irritation. In addition, the above terms may be used interchangeably with the terms ‘for diagnosing changes in skin brightness/irritation due to hydroquinone’, ‘for diagnosing changes in skin brightness due to hydroquinone/diagnosing the degree of skin irritation’, and ‘for determining changes in skin brightness due to hydroquinone/diagnosing the degree of skin irritation’. You can.

In the present invention, the term “prescription” may refer to a substance having a specific composition, and is a concept that may further include a recommended method of use or a recommended lifestyle along with a single or multiple substances. Taking the present invention as an example, a “prescription” may include a prescription for hydroquinone alone, a prescription for hydroquinone plus one or more additional substances, such as substances that induce increased skin brightness, skin irritation relievers, agent stabilizing agents, and agents. It may be one or more selected from the group consisting of texture crystal materials.

Primers used for amplification of the polymorphic marker include template-directed DNA under appropriate conditions (e.g., four different nucleoside triphosphates and DNA, a polymerizer such as RNA polymerase or reverse transcriptase) in an appropriate buffer and appropriate temperature. It refers to a single-stranded oligonucleotide that can serve as a starting point for synthesis. The appropriate length of the primer may vary depending on the purpose of use, but is usually 15 to 30 nucleotides. Short primer molecules generally require lower temperatures to form stable hybrids with the template. The primer sequence need not be completely complementary to the template, but should be sufficiently complementary to hybridize to the template.

In the present invention, the term "primer" is a base sequence having a short free 3' terminal hydroxyl group, which can form a base pair with a complementary template and serves as a starting point for copying the template strand. It refers to a short sequence that functions as a point. Primers can initiate DNA synthesis in the presence of four different nucleoside triphosphates and reagents for polymerization (i.e., DNA polymerase or reverse transcriptase) in an appropriate buffer and temperature. By performing PCR amplification, skin type can be predicted through the level of production of the desired product. PCR conditions and lengths of sense and antisense primers can be modified based on those known in the art.

The probe or primer of the present invention can be chemically synthesized using the phosphoramidite solid support method or other well-known methods. These nucleic acid sequences can also be modified using many means known in the art. Non-limiting examples of such modifications include methylation, “capsation,” substitution of a native nucleotide with one or more homologs, and modifications between nucleotides, such as uncharged linkages (e.g., methyl phosphonates, phosphotriesters, phosphoroamidate, carbamate, etc.) or charged linkages (e.g. phosphorothioate, phosphorodithioate, etc.).

Another aspect of the present invention provides a kit for determining hydroquinone skin brightness change/irritation, including the composition for determining hydroquinone skin brightness change/irritation. The kit may be an RT-PCR kit or a DNA chip kit, but is not limited thereto.

The kit of the present invention can diagnose skin type by confirming the SNP polymorphism marker, which is a marker for determining skin type, through amplification, or by checking the expression level of mRNA for the expression level of the SNP polymorphism marker. As a specific example, in the present invention, a kit for measuring the mRNA expression level of a marker for determining skin brightness change/stimulation by hydroquinone may be a kit containing the essential elements required to perform RT-PCR. In addition to each pair of primers specific for the gene of the marker for determining skin brightness change/irritation by hydroquinone, the RT-PCR kit contains a test tube or other suitable container, reaction buffer (pH and magnesium concentration vary) , deoxynucleotides (dNTPs), enzymes such as Taq-polymerase and reverse transcriptase, DNase, RNAse inhibitor, DEPC-water, sterilized water, etc. It may also include a pair of primers specific to the gene used as a quantitative control. Also specifically, the kit of the present invention may be a kit for determining skin type that includes the essential elements necessary to perform a DNA chip. DNA chip kits are generally made by attaching nucleic acid species in a gridded array to a flat solid support plate, typically a glass surface no larger than a microscope slide. Nucleic acids are arranged uniformly on the chip surface, forming a DNA chip. It is a tool that enables massively parallel analysis by causing multiple hybridization reactions between the nucleic acids on the chip and the complementary nucleic acids contained in the solution treated on the chip surface.

Another aspect of the present invention provides a microarray for determining changes in hydroquinone skin brightness, including the composition for determining changes in hydroquinone skin brightness.

Another aspect of the present invention provides a microarray for determining hydroquinone skin irritation including the composition for determining hydroquinone skin irritation.

The microarray may include DNA or RNA polynucleotides. The microarray consists of a conventional microarray except that the probe polynucleotide includes the polynucleotide of the present invention.

Methods for manufacturing microarrays by immobilizing probe polynucleotides on a substrate are well known in the art. The probe polynucleotide refers to a polynucleotide capable of hybridization, and refers to an oligonucleotide capable of sequence-specific binding to the complementary strand of a nucleic acid. The probe of the present invention is an allele-specific probe, in which a polymorphic site is present among nucleic acid fragments derived from two members of the same species, and hybridizes to the DNA fragment derived from one member, but does not hybridize to the fragment derived from the other member. . In this case, hybridization conditions must be sufficiently stringent to ensure that only one of the alleles hybridizes, as there is a significant difference in hybridization strength between alleles. This can lead to good hybridization differences between different allelic forms. The probe of the present invention can be used in a method for detecting alleles and diagnosing changes in skin brightness and/or irritation caused by hydroquinone. The diagnostic method includes detection methods based on hybridization of nucleic acids such as Southern blot, and in the method using a DNA chip, it may be provided in a form pre-bound to the substrate of the DNA chip. The hybridization can usually be performed under strict conditions, for example, at a salt concentration of 1M or less and a temperature of 25°C or more. For example, conditions of 5x SSPE (750mM NaCl, 50mM Na Phosphate, 5mM EDTA, pH 7.4) and 25-30°C may be suitable for allele-specific probe hybridization.

The process of immobilizing the probe polynucleotide associated with the skin diagnosis of the present invention on a substrate can also be easily manufactured using this conventional technology. Additionally, hybridization of nucleic acids and detection of hybridization results on microarrays are well known in the art. The detection may be performed, for example, by labeling a nucleic acid sample with a labeling material capable of generating a detectable signal, including fluorescent substances such as Cy3 and Cy5, and then hybridizing it on a microarray and generating a detectable signal from the labeling material. Hybridization results can be detected by detecting the signal.

Another aspect of the present invention includes the steps of (a) amplifying the polymorphic site of the single nucleotide polymorphism marker in DNA obtained from a sample isolated from an individual or hybridizing it with a probe; and (b) confirming the base of the amplified or hybridized polymorphic site in step (a).

Another aspect of the present invention includes the steps of (a) amplifying the polymorphic site of the single nucleotide polymorphism marker in DNA obtained from a sample isolated from an individual or hybridizing it with a probe; and (b) confirming the base of the amplified or hybridized polymorphic site in step (a). It provides a method of providing information on determining skin irritation by hydroquinone.

As used herein, the term “subject” refers to a subject for diagnosing skin irritation caused by hydroquinone. DNA can be obtained from samples such as hair, urine, blood, various body fluids, separated tissues, separated cells, or saliva, but is not limited thereto.

Any method known to those skilled in the art can be used to obtain genomic DNA in step (a).

The step of amplifying the polymorphic site of the single nucleotide polymorphism marker from the DNA obtained in step (a) or hybridizing it with a probe can be performed by any method known to those skilled in the art. For example, target nucleic acid can be amplified through PCR and purified. In addition, ligase chain reaction (LCR) (Wu and Wallace, Genomics 4, 560 (1989), Landegren et al., Science 241, 1077 (1988)), transcription amplification (Kwoh et al., Proc. Natl. Acad. Sci. USA 86, 1173 (1989)) and self-sustaining sequence cloning (Guatelli et al., Proc. Natl. Acad. Sci. USA 87, 1874 (1990)) and nucleic acid-based sequence amplification (NASBA) can be used.

Among the above methods, determining the base of the polymorphic site in step (b) includes sequencing analysis, hybridization by microarray, allele specific PCR, and dynamic allele-specific hybridization. DASH), PCR extension assay, SSCP, PCR-RFLP assay or TaqMan technique, SNPlex platform (Applied Biosystems), mass spectrometry (e.g., Sequenom's MassARRAY system), mini-sequencing method, Bio-Plex Systems (BioRad), CEQ and SNPstream systems (Beckman), Molecular Inversion Probe array technology (e.g., Affymetrix GeneChip), and BeadArray Technologies (e.g., Illumina GoldenGate and Infinium assays). By the above methods or other methods available to those skilled in the art to which the present invention pertains, one or more alleles in a polymorphic marker, including microsatellites, SNPs or other types of polymorphic markers, can be identified. Determining the base of such a polymorphic site can be specifically performed using a SNP chip.

The method additionally applies (c) when the base of the amplified or hybridized polymorphic site includes one or more bases that are minor alleles according to the single nucleotide polymorphism marker, change in skin brightness and/or skin brightness due to hydroquinone The level of stimulation may be judged to be high or low, but is not limited to this.

In the present invention, the term “SNP chip” refers to a DNA microarray that can identify each base of hundreds of thousands of SNPs at once.

The TaqMan method includes (1) designing and producing primers and TaqMan probes to amplify the desired DNA fragment; (2) labeling probes of different alleles with FAM dye and VIC dye (Applied Biosystems); (3) using the DNA as a template and performing PCR using the primers and probes; (4) After the above PCR reaction is completed, analyzing and confirming the TaqMan assay plate with a nucleic acid analyzer; and (5) determining the genotypes of the polynucleotides of step (1) from the analysis results.

In the above, sequencing analysis can be performed using a conventional method for determining a base sequence, or can be performed using an automated genetic analyzer. In addition, allele-specific PCR refers to a PCR method that amplifies the DNA fragment where the SNP is located using a primer set including a primer designed with the base where the SNP is located as the 3' end. The principle of the method is, for example, when a specific base is substituted from A to G, PCR is performed by designing a primer containing A as the 3' terminal base and a reverse primer capable of amplifying a DNA fragment of an appropriate size. When performing the reaction, if the base at the SNP position is A, the amplification reaction is performed normally and a band at the desired position is observed, and if the base is replaced by G, the primer can complement the template DNA, but 3 ' It takes advantage of the fact that the amplification reaction is not performed properly due to the failure of complementary bonds at the terminal end. DASH can be performed by conventional methods, and specifically by the method by Prince et al.

Meanwhile, in the PCR extension analysis, a DNA fragment containing the base where the single nucleotide polymorphism is located is first amplified with a pair of primers, and then all nucleotides added to the reaction are inactivated by dephosphorylation, followed by SNP-specific extension primers, This is accomplished by performing a primer extension reaction by adding dNTP mixture, dideoxynucleotide, reaction buffer, and DNA polymerase. At this time, the extension primer uses the base immediately adjacent in the 5' direction of the base where the SNP is located as the 3' end, and the dNTP mixture excludes nucleic acids having the same base as the dideoxynucleotide, and the dideoxynucleotide represents the SNP. It is selected from one of the base types. For example, when there is a substitution from A to G and a mixture of dGTP, dCTP and TTP and ddATP are added to the reaction, the primer is extended by DNA polymerase at the base where the substitution occurred, and after a few bases, A The primer extension reaction is terminated by ddATP at the position where the base first appears. If the substitution has not occurred, the extension reaction is terminated at that position, so that the type of base representing the SNP can be determined by comparing the lengths of the extended primers.

At this time, the detection method is to detect the SNP by detecting fluorescence using a genetic analyzer (for example, ABI's Model 3700, etc.) used for general base sequence determination in the case of fluorescently labeled extension primers or dideoxynucleotides. When using unlabeled extension primers and dideoxynucleotides, the SNP can be detected by measuring the molecular weight using matrix assisted laser desorption ionization-time of flight (MALDI-TOF) technique.

As one aspect for achieving the object of the present invention, the present invention provides a composition for anti-irritation by hydroquinone containing an irritation reliever as an active ingredient.

For the purpose of the present invention, the irritation reliever may be a Western red pepper herb extract or Salvia Salvia root extract, but is not limited thereto.

The term 'Hypericum perforatum' is a perennial herb of the family Asteraceae of the order Plasmodium order. It is one of the herbs native to Europe and Western Asia. It is a plant that contains an ingredient that causes depression, and the ingredient hypericin is known to increase serotonin levels.

The term 'salvia miltiorrhiza' is a perennial herb belonging to the Lamiaceae family. It is native to China and is also called powdered root or blood root. The roots are red, have a peculiar smell, and taste slightly bitter. Salvian ginseng root is used as a medicine in oriental medicine, and salvianolic acids A and B in the extract are known to be effective in blood circulation and hyperlipidemia.

In the present invention, “extract” refers to a liquid component obtained by immersing the desired substance in various solvents and then extracting the substance for a certain period of time at room temperature, low temperature, or heated state, and removing the solvent from the liquid component. It refers to the result of solid content, etc. In addition, in addition to the above results, it can be comprehensively interpreted to include all dilutions of the results, concentrates thereof, crude preparations, purifications, etc. of the above results.

In the present invention, the extract can be extracted from various organs of natural, hybrid, and mutated plants, for example, roots, aerial parts, stems, leaves, flowers, fruit bodies, fruit peels, as well as plant tissues. do. Additionally, the extract can be obtained by extraction with water or various organic solvents. At this time, the organic solvent used is not particularly limited as long as it can obtain an extract, but may specifically be water, a polar solvent, or a non-polar solvent, and more specifically, water, a lower alcohol having 1 to 6 carbon atoms ( methanol, ethanol, propanol, or butanol, etc.), or a mixed solvent thereof, and most specifically, it may be methanol or a mixed solvent thereof.

In addition, the method for obtaining the extract is not particularly limited as long as it is possible to obtain the Western red pepper herb extract or Salvia Salvia Salvia root extract, but specifically, the roots, stems, leaves, fruits, and flowers of the Western red pepper herb extract or Salvia Salvia Militias root extract. , these dried products, processed products, etc. can be immersed in the above solvent and extracted at room temperature using methods such as cold immersion extraction, heat extraction, ultrasonic extraction by applying ultrasonic waves, and reflux extraction using a reflux condenser.

The irritation reliever of the present invention may be contained in an amount of about 0.1 to 30% by weight based on the total weight of the composition, but is not limited thereto. As an example, the irritation reliever of the present invention may be used in an amount of about 0.1 to 30% by weight, about 1 to 30% by weight, about 1 to 25% by weight, about 1 to 20% by weight, or about 1 to 15% by weight, based on the total weight of the composition. , about 1 to 10 weight percent, or about 1 to 5 weight percent.

For the purpose of the present invention, the irritation reliever has the effect of weakening the irritation caused by hydroquinone.

In the present invention, the term “about” may appear before a specific numeric value. As used in this application, the term “about” includes not only the exact number written after the term, but also a range that is approximately that number or close to that number. By considering the context in which the number is presented, one can determine whether it is close to or close to the specific number mentioned. As an example, the term “about” may refer to a range of -10% to +10% of a numeric value. As another example, the term “about” may refer to a range of -5% to +5% of a given numeric value. However, it is not limited to this.

The composition is characterized as being personalized. Specifically, a personalized irritant reliever can be provided to individuals containing a single nucleotide polymorphism (SNP) marker for diagnosing sensitivity to specific hydroquinone.

As an example, the single nucleotide polymorphism (SNP) marker for determining hydroquinone stimulation may be one or more selected from Tables 2 and 3, but is not limited thereto.

The gene for determining hydroquinone irritation can provide information for selecting a specific irritation reliever for each individual. As a reference example, in the case of an individual with the PRKCE gene, the irritation caused by hydroquinone can be alleviated when treated with an irritant extract and/or Salvia Salvia root extract, but optimal stimulation is determined depending on the genetic characteristics of each individual. Combinations of palliative agents may be provided.

This means that it is possible to select specific single nucleotide polymorphism (SNP) markers that have a significant correlation with irritation caused by hydroquinone and provide a personalized irritation reliever that can help relieve irritation caused by hydroquinone based on genetic information. It suggests.

The composition of the present invention can be used as a cosmetic composition and can be formulated in various forms. The cosmetic composition according to the present invention includes solution, external ointment, cream, foam, nourishing lotion, softening lotion, pack, softening water, emulsion, makeup base, essence, soap, liquid cleanser, bath agent, sunscreen cream, sun oil, suspension, Can be prepared in a formulation selected from the group consisting of emulsions, pastes, gels, lotions, powders, soaps, surfactant-containing cleansers, oils, powder foundations, emulsion foundations, wax foundations, patches and sprays. That is not the case.

In addition, the cosmetic composition of the present invention may additionally include one or more cosmetically acceptable carriers that are blended with general skin cosmetics, and common ingredients include, for example, oil, water, surfactant, moisturizer, lower alcohol, Thickeners, chelating agents, pigments, preservatives, fragrances, etc. may be appropriately mixed, but are not limited thereto.

Cosmetically acceptable carriers included in the cosmetic composition of the present invention vary depending on the formulation.

When the formulation of the present invention is an ointment, paste, cream or gel, the carrier ingredient may be animal oil, vegetable oil, wax, paraffin, starch, tracant, cellulose derivative, polyethylene glycol, silicone, bentonite, silica, talc, zinc oxide or Mixtures of these can be used.

When the formulation of the present invention is a powder or spray, lactose, talc, silica, aluminum hydroxide, calcium silcate, polyamide powder, or mixtures thereof may be used as the carrier ingredient, and especially in the case of a spray, chloro May contain propellants such as fluorohydrocarbons, propane/butane or dimethyl ether.

When the formulation of the present invention is a solution or emulsion, a solvent, solubilizing agent or emulsifying agent is used as a carrier component, such as water, ethanol, isopropanol, ethyl carbonate, ethyl acetate, benzyl alcohol, benzyl benzoate, propylene glycol, 1,3-Butylglycol oil may be used, in particular cottonseed oil, peanut oil, corn germ oil, olive oil, castor oil and sesame oil, glycerol aliphatic esters, polyethylene glycol or fatty acid esters of sorbitan. there is.

When the formulation of the present invention is a suspension, the carrier ingredients include water, a liquid diluent such as ethanol or propylene glycol, a suspending agent such as ethoxylated isostearyl alcohol, polyoxyethylene sorbitol ester, and polyoxyethylene sorbitan ester, and miso. Crystalline cellulose, aluminum metahydroxide, bentonite, agar, or tracant may be used.

When the formulation of the present invention is soap, alkali metal salts of fatty acids, fatty acid hemiester salts, fatty acid protein hydrolyzates, isethionates, lanolin derivatives, fatty alcohols, vegetable oils, glycerol, sugars, etc. may be used as carrier ingredients. You can.

In another aspect, the composition provides a pharmaceutical composition for preventing or treating irritation caused by hydroquinone.

The pharmaceutical composition of the present invention can be used as a single preparation, or can be prepared and used as a combination preparation by additionally containing a drug known to have an irritation-relieving effect, and is formulated using a pharmaceutically acceptable carrier or excipient to form a unit dosage form. It can be manufactured or manufactured by placing it in a multi-capacity container.

The pharmaceutical composition can be administered in a pharmaceutically effective amount.

The pharmaceutical composition of the present invention may be administered in a pharmaceutically effective amount, wherein the term "pharmaceutically effective amount" is sufficient to treat or prevent a disease with a reasonable benefit/risk ratio applicable to medical treatment or prevention. It refers to the amount, and the effective dose level is the severity of the disease, the activity of the drug, the patient's age, weight, health, gender, the patient's sensitivity to the drug, the administration time, administration route, and excretion rate of the composition of the present invention used. Treatment period , may be determined according to factors including the composition of the present invention and the drugs used in combination or simultaneous use, and other factors well known in the medical field. The pharmaceutical composition of the present invention can be administered alone or in combination with a known pterygium treatment agent. It is important to consider all of the above factors and administer the amount that will achieve maximum effect with the minimum amount without side effects.

In another aspect, the composition provides a quasi-drug composition for alleviating or improving irritation caused by hydroquinone.

In the present invention, the term "quasi-drug" refers to products that have a milder effect than pharmaceuticals among products used for the purpose of diagnosing, treating, improving, alleviating, treating, or preventing diseases in humans or animals. For example, according to the Pharmaceutical Affairs Act of the Republic of Korea, Quasi-drugs exclude products used for medicinal purposes and include products used to treat or prevent diseases in humans and animals, and products that have a mild or no direct effect on the human body.

In one embodiment, the quasi-drug composition of the present invention can be manufactured in a form selected from the group consisting of body cleanser, shampoo, conditioner, foam, soap, mask, ointment, cream, lotion, essence, and spray, but is limited thereto. no.

In another aspect, the present invention includes the steps of (a) obtaining a biological sample from an individual; (b) amplifying a polymorphic site of one or more single nucleotide polymorphism markers selected from the group consisting of rs17033965, rs11898074, rs74684800, rs12712955, rs2711292, rs1562653, and rs12712969 from the biological sample or hybridizing with a probe; and (c) identifying the base of the amplified or hybridized polymorphic site in step (b). It provides an information providing method for selecting a customized irritation reliever for alleviating irritation caused by hydroquinone.

Additionally, by confirming the base of the amplified or hybridized polymorphic site in step (c), a personalized irritation reliever can be selected to relieve irritation caused by hydroquinone.

The irritation reliever may be a Western red pepper herb extract or Salvia Salvia root extract, but is not limited thereto.

A method according to an aspect of the present invention includes the steps of (a) inputting SNP information of a prediction target entity to a prediction terminal, (b) the prediction terminal applying the skin brightness change prediction model previously stored in the prediction terminal or server. Querying SNP information to calculate predicted skin brightness change information, and outputting the calculated predicted skin brightness change information, wherein the skin brightness change prediction model is configured to allow a learning processor to apply SNP information and an object having the SNP information. When a substance containing hydroquinone is applied, first skin brightness change learning data consisting of a pair of skin brightness change information are learned, and when SNP information is queried, a substance containing hydroquinone is applied to an individual with the queried SNP information. Provides a method for predicting skin brightness change in a prescription containing hydroquinone, which is a first skin brightness change prediction model that outputs first predicted skin brightness change information.

In another aspect of the present invention, the SNPs marker selection device further includes the step of selecting one or more SNPs related to a change in skin brightness upon application of hydroquinone from among a plurality of SNPs included in the SNP information, and SNP information of the learning data includes at least some of the one or more SNPs selected by the SNPs marker selection device, and may not include SNPs not selected by the SNPs marker selection device.

In another aspect of the present invention, one or more SNPs related to the change in skin brightness may be any one or more selected from Table 1.

In another aspect of the present invention, as the learning process learns the learning data, a weight that affects the change in skin brightness when applying a substance containing hydroquinone can be determined for each SNP included in the SNP information, and the weight is , it may be calculated using a first value determined according to the number of major alleles and minor alleles included in each SNP, and a second value whose value changes through learning. there is.

In another aspect of the invention, the first value is: major allele - major allele, major allele - minor allele, and minor allele - SNPs type of the minor allele. Each can be given a different value.

In another aspect of the present invention, the first value may have a higher value among the SNPs types with a greater number of lower-level alleles.

In another aspect of the present invention, the first value may be a value that does not change even if learning occurs.

In another aspect of the present invention, the weight is a value obtained by multiplying the first value and the second value, and the predicted skin brightness change information may be calculated by summing the weights for each SNP.

In another aspect of the present invention, the skin brightness change prediction model is such that the learning processor provides SNP information, one or more prescription information among a plurality of prescriptions including a prescription containing hydroquinone, and predicts a change in skin brightness to an individual having the SNP information according to the prescription. By learning the second skin brightness change learning data consisting of pairs of skin brightness change information when applying the substance, when SNP information is queried, when each substance according to a plurality of prescriptions is applied to an individual with the queried SNP information It may be a second skin brightness change prediction model that outputs the second predicted skin brightness change information.

In another aspect of the present invention, the skin brightness change prediction model includes a learning processor having, in addition to SNP information, one or more of age information, gender information, lifestyle information, environmental information, and skin characteristic information, and the SNP information. When a substance containing hydroquinone is applied to an object, the third skin brightness change learning data consisting of pairs of skin brightness change information are learned, and among SNP information, age information, gender information, lifestyle information, environmental information, and skin characteristic information When one or more information is queried, it may be a third skin brightness change prediction model that outputs third predicted skin brightness change information when a substance containing hydroquinone is applied to an entity having the queried information.

In another aspect of the present invention, the skin brightness change prediction model is such that the learning processor, in addition to SNP information, provides a prescription including one or more of age information, gender information, lifestyle information, environmental information, and skin characteristic information, and hydroquinone. By learning the fourth skin brightness change learning data consisting of a pair of prescription information of one or more prescriptions including a plurality of prescriptions and skin brightness change information when a substance according to the prescription is applied to an object having the SNP information, SNP information and when one or more of age information, gender information, lifestyle information, environmental information, and skin characteristic information is queried, a fourth predicted change in skin brightness when each of a plurality of prescription substances is applied to an individual having the queried information. It may be a fourth skin brightness change prediction model that outputs information.

In another aspect of the present invention, step (b) includes the step (b) of the prediction terminal querying the skin brightness change prediction model for the SNP information to calculate predicted skin brightness change information for each of a plurality of prescriptions. It may further include (c) a step of the prescription selection unit selecting one or more prescriptions among a plurality of prescriptions using the predicted skin brightness change information.

In another aspect of the present invention, step (c) may further include the step of the prescription selection unit 33 selecting one or more prescriptions with a large value of the predicted skin brightness change information from among a plurality of prescriptions.

In another aspect of the present invention, the plurality of prescriptions include a hydroquinone prescription and a prescription in which hydroquinone further contains an additional substance, wherein the additional substance includes a skin brightening increase inducing agent, a skin irritation reliever, a formulation stabilizing agent, and a formulation texture determining agent. It may be one or more selected from a group consisting of substances.

In another aspect of the present invention, in step (b), the prediction terminal queries the SNP information to a skin irritation prediction model previously stored in the prediction terminal or the server, and the calculated information is additionally output, or the prediction terminal Further comprising the step of additionally outputting a critical irritation score determined in a threshold irritation score prediction model pre-stored in the terminal or the server, wherein the skin irritation prediction model is configured to: By learning the first stimulus learning data consisting of a pair of stimulus information when a substance containing It is a model that calculates predicted stimulation information, and the critical stimulation score prediction model is a first pair consisting of a stimulation score predetermined by the learning processor according to SNP information and the presence or absence of stimulation when hydroquinone is applied to an individual having the SNP information. 1 Threshold stimulus learning data may be learned to determine the threshold stimulus score that causes skin irritation when a substance containing hydroquinone is applied.

In another aspect of the present invention, according to a comparison result between the output critical stimulation score and the stimulation score determined according to the queried SNP information, the agent when a substance containing hydroquinone is applied to an individual having the queried SNP information 2 The step of calculating the predicted stimulus information may be further included.

In another aspect of the present invention, (c) the prescription selection unit queries among a plurality of prescriptions containing hydroquinone using one or more of the predicted skin brightness change information, the first predicted stimulus information, and the second predicted stimulus information. A step of selecting one or more prescriptions suitable to be prescribed to an individual having the SNP information may be further included.

In addition, one aspect of the present invention predicts skin brightness change by learning first skin brightness change learning data consisting of a pair of SNP information and skin brightness change information when a substance containing hydroquinone is applied to an individual having the SNP information. A server that generates a model, when SNP information is queried, generates a first skin brightness change prediction model that outputs first predicted skin brightness change information when a substance containing hydroquinone is applied to an object having the queried SNP information. A prediction terminal that receives SNP information of a server and a prediction target object and outputs first predicted skin brightness change information calculated by querying the SNP information input to the first skin brightness change prediction model generated by the server. , provides a system for predicting changes in skin brightness of prescriptions containing hydroquinone.

In another aspect of the present invention, it further includes a SNPs marker selection device for selecting one or more SNPs related to changes in skin brightness upon application of hydroquinone among the plurality of SNPs included in the SNP information, and the SNP information of the learning data includes the above At least some of the one or more SNPs selected by the SNPs marker selection device may be included, and SNPs not selected by the SNPs marker selection device may not be included.

In another aspect of the present invention, as the running process of the server learns the learning data, a weight that affects the change in skin brightness when applying a substance containing hydroquinone can be determined for each SNP included in the SNP information, The weight is calculated using a first value determined according to the number of major alleles and minor alleles included in each SNP and a second value whose value changes through learning. It may be.

In another aspect of the present invention, the first value is: major allele - major allele, major allele - minor allele, and minor allele - SNPs type of the minor allele. Each can be given a different value.

In another aspect of the present invention, the skin brightness change prediction model is such that the server provides SNP information, one or more prescription information among a plurality of prescriptions including a prescription containing hydroquinone, and a substance according to the prescription to an entity having the SNP information. By learning the second skin brightness change learning data consisting of pairs of skin brightness change information when applied, when SNP information is queried, when each substance according to a plurality of prescriptions is applied to an individual with the queried SNP information It may be a second skin brightness change prediction model that outputs second predicted skin brightness change information.

In another aspect of the present invention, the skin brightness change prediction model is such that the server has, in addition to SNP information, one or more of age information, gender information, lifestyle information, environmental information, and skin characteristic information, and the SNP information. When a substance containing hydroquinone is applied to an object, the third skin brightness change learning data consisting of pairs of skin brightness change information are learned, and among SNP information, age information, gender information, lifestyle information, environmental information, and skin characteristic information When one or more information is queried, it may be a third skin brightness change prediction model that outputs third predicted skin brightness change information when a substance containing hydroquinone is applied to an object having the queried information.

In another aspect of the present invention, the prediction terminal queries the skin brightness change prediction model for the SNP information to calculate predicted skin brightness change information for each of a plurality of prescriptions, and the system calculates the predicted skin brightness change information Using the method, it may further include a prescription selection unit that selects one or more prescriptions from a plurality of prescriptions.

In another aspect of the present invention, the prescription selection unit may select one or more prescriptions with a large value of the predicted skin brightness change information from among a plurality of prescriptions.

In another aspect of the present invention, the prediction terminal further outputs information calculated by querying the SNP information input to the skin irritation prediction model generated by the server, or determines the threshold irritation score prediction model generated by the server. The score is additionally output, and the skin irritation prediction model learns first stimulus learning data consisting of pairs of stimulus information when the learning processor applies SNP information and a substance containing hydroquinone to an object having the SNP information. Thus, when SNP information is queried, it is a model that calculates the first predicted stimulus information when a substance containing hydroquinone is applied to an object having the queried SNP information, and the threshold stimulus score prediction model is where the learning processor calculates the SNP According to the information, first critical stimulus learning data consisting of a predetermined stimulation score and a pair of stimulation presence or absence when hydroquinone is applied to an entity with the SNP information are learned, causing skin irritation when a substance containing hydroquinone is applied. The threshold stimulus score may be output.

In another aspect of the present invention, the prediction terminal is configured to provide a substance containing hydroquinone to an entity having the queried SNP information according to a comparison result between the output critical stimulus score and the stimulus score determined according to the queried SNP information. The second predictive stimulus information when applied can be further calculated.

In another aspect of the present invention, using the skin brightness change information and one or more of the first predictive stimulus information and the second predictive stimulus information, among a plurality of prescriptions containing hydroquinone, an individual having the queried SNP information It may further include a prescription selection unit that selects one or more prescriptions suitable to be prescribed.

One aspect of the present invention provides a computer program, stored on a computer-readable recording medium, for executing the above-described method.

Meanwhile, a method according to another aspect of the present invention includes the steps of (a) inputting SNP information of a prediction target entity to a prediction terminal, and (b) the prediction terminal using a skin irritation prediction model previously stored in the prediction terminal or server. Comprising the step of querying the SNP information to calculate predicted stimulation information, and outputting the calculated predicted stimulation information, wherein the skin irritation prediction model is configured to include hydroquinone in an object having SNP information and the SNP information. By learning first stimulus learning data consisting of pairs of stimulus information when the material is applied, when SNP information is queried, first predicted stimulus information when a material containing hydroquinone is applied to an individual with the queried SNP information Provides a method for predicting skin irritation of a prescription containing hydroquinone, which is a first irritation prediction model that outputs.

In another aspect of the present invention, the SNPs marker selection device further includes the step of selecting one or more SNPs related to the degree of skin irritation upon application of hydroquinone from among a plurality of SNPs included in the SNP information, and SNP information of the learning data includes at least some of the one or more SNPs selected by the SNPs marker selection device, and may not include SNPs not selected by the SNPs marker selection device.

In another aspect of the present invention, one or more SNPs related to the degree of skin irritation may be any one or more selected from Table 2.

In another aspect of the present invention, as the learning process learns the learning data, a weight that affects skin irritation when applying a substance containing hydroquinone can be determined for each SNP included in the SNP information, and the weight is, It may be calculated using a first value determined according to the number of major alleles and minor alleles included in each SNP, and a second value whose value changes through learning. .

In another aspect of the present invention, the weight is a value obtained by multiplying the first value and the second value, and the predictive stimulus information may be calculated by summing the weights for each SNP.

In another aspect of the present invention, the skin irritation prediction model is configured to provide the learning processor with SNP information, one or more prescription information from a plurality of prescriptions including a prescription containing hydroquinone, and a substance according to the prescription to an entity having the SNP information. By learning the second stimulus learning data consisting of pairs of stimulus information when applied, when SNP information is queried, a second prediction stimulus when each of the substances according to a plurality of prescriptions is applied to an individual with the queried SNP information It may be a second stimulus prediction model that outputs information.

In another aspect of the present invention, the skin irritation prediction model includes a learning processor, in addition to SNP information, one or more of age information, gender information, lifestyle information, environmental information, and skin characteristic information, and an individual having the SNP information. When a substance containing hydroquinone is applied, third stimulus learning data consisting of pairs of stimulus information are learned, and one or more of SNP information, age information, gender information, lifestyle information, environmental information, and skin characteristic information is queried. If so, it may be a third stimulation prediction model that outputs third prediction stimulation information when a substance containing hydroquinone is applied to an entity having the queried information.

In another aspect of the present invention, the skin irritation prediction model is such that the learning processor includes, in addition to SNP information, one or more of age information, gender information, lifestyle information, environmental information, and skin characteristic information, and a prescription including hydroquinone. By learning the fourth stimulus learning data consisting of a pair of prescription information of one or more prescriptions among a plurality of prescriptions and stimulus information when the substance according to the prescription is applied to an object having the SNP information, SNP information, age information, and gender When one or more of information, lifestyle information, environmental information, and skin characteristic information is queried, a fourth stimulus that outputs fourth predictive stimulus information when each of a plurality of prescription substances is applied to an entity having the queried information. It may be a predictive model.

In another aspect of the present invention, step (b) further includes the step of (b) the prediction terminal querying the skin irritation prediction model for the SNP information to calculate predicted stimulation information for each of a plurality of prescriptions, , (c) a prescription selection unit may further include the step of selecting one or more prescriptions among a plurality of prescriptions using the predicted stimulus information.

In another aspect of the present invention, step (c) may further include the step of the prescription selection unit 33 selecting one or more prescriptions with a small value of the predictive stimulus information from among a plurality of prescriptions.

In another aspect of the present invention, in step (b), the prediction terminal queries the SNP information to a skin brightness change prediction model pre-stored in the prediction terminal or server to calculate predicted skin brightness change information, and calculates It further includes the step of outputting predicted skin brightness change information, wherein the learning processor learns the first skin brightness change consisting of a pair of SNP information and skin brightness change information when a substance containing hydroquinone is applied to an object having the SNP information. By learning the data, when SNP information is queried, predicted skin brightness change information when a substance containing hydroquinone is applied to an individual with the queried SNP information may be additionally output.

In another aspect of the present invention, (c) a prescription selection unit uses one or more of the first predicted stimulus information and the predicted skin brightness change information to select an individual with the queried SNP information among a plurality of prescriptions containing hydroquinone. The step of selecting one or more prescriptions suitable to be prescribed may be further included.

In addition, one aspect of the present invention is to generate a skin irritation prediction model by learning first stimulus learning data consisting of a pair of SNP information and stimulus information when a substance containing hydroquinone is applied to an entity having the SNP information. As a server, when SNP information is queried, a server that generates a first stimulation prediction model that outputs first prediction stimulation information when a substance containing hydroquinone is applied to an entity having the queried SNP information, and the SNP of the entity to be predicted. A skin irritation prediction system for a prescription containing hydroquinone, including a prediction terminal that receives information and outputs first prediction stimulation information calculated by querying the SNP information input to the first stimulation prediction model generated by the server. to provide.

In another aspect of the present invention, it further includes a SNPs marker selection device for selecting one or more SNPs related to the degree of skin irritation upon application of hydroquinone among the plurality of SNPs included in the SNP information, and the SNP information of the learning data includes the At least some of the one or more SNPs selected by the SNPs marker selection device may be included, and SNPs not selected by the SNPs marker selection device may not be included.

In another aspect of the present invention, as the learning process of the server learns the learning data, a weight that affects skin irritation when applying a substance containing hydroquinone can be determined for each SNP included in the SNP information, and the weight is calculated using a first value determined according to the number of major alleles and minor alleles included in each SNP and a second value whose value changes through learning. You can.

In another aspect of the present invention, the skin irritation prediction model is such that the server provides SNP information, one or more prescription information among a plurality of prescriptions including a prescription containing hydroquinone, and a substance according to the prescription to an entity having the SNP information. By learning second stimulus learning data consisting of pairs of stimulus information when applied, when SNP information is queried, second predicted stimulus information when each substance according to a plurality of prescriptions is applied to an individual having the queried SNP information It may be a second stimulus prediction model that outputs.

In another aspect of the present invention, the skin irritation prediction model is such that, in addition to SNP information, the server provides one or more of age information, gender information, lifestyle information, environmental information, and skin characteristic information, and an individual having the SNP information. By learning the third stimulus learning data consisting of a pair of stimulus information when a substance containing hydroquinone is applied, one or more of SNP information, age information, gender information, lifestyle information, environmental information, and skin characteristic information is provided. When queried, it may be a third stimulation prediction model that outputs third prediction stimulation information when a substance containing hydroquinone is applied to an entity having the queried information.

In another aspect of the present invention, the prediction terminal queries the skin irritation prediction model for the SNP information to calculate prediction stimulation information for each of a plurality of prescriptions, and the system uses the prediction stimulation information to generate a plurality of It may further include a prescription selection unit that selects one or more prescriptions among the prescriptions.

In another aspect of the present invention, the prescription selection unit may select one or more prescriptions with a small value of the predicted skin irritation information from among a plurality of prescriptions.

In another aspect of the present invention, the prediction terminal queries the SNP information input to the skin brightness change prediction model generated by the server and outputs the calculated predicted skin brightness change information, and the skin brightness change prediction model is, The learning processor learns first skin brightness change learning data consisting of a pair of SNP information and skin brightness change information when a substance containing hydroquinone is applied to an object having the SNP information, and when SNP information is queried, the queried SNP Information on changes in predicted skin brightness may be additionally output when a substance containing hydroquinone is applied to an object having the information.

In another aspect of the present invention, one or more prescriptions suitable to be prescribed to an individual with the queried SNP information among a plurality of prescriptions containing hydroquinone, using one or more of the first predicted stimulus information and the predicted skin brightness change information. It may further include a prescription selection unit for selecting.

Meanwhile, a method according to another aspect of the present invention includes the steps of (a) inputting SNP information of a prediction target entity to a prediction terminal, and (b) the prediction terminal using a threshold stimulus score prediction model previously stored in the prediction terminal or server. Comparing the threshold stimulation score determined in and the stimulation score determined according to the SNP information, and predicting stimulation information when hydroquinone is applied to an individual with the input SNP information according to the comparison result, and the threshold The stimulation score prediction model is a stimulation score determined by the learning processor according to the number of SNPs and SNPs type included in the SNP information, and the presence or absence of stimulation when a substance containing hydroquinone is applied to an object having the SNP information. Provides a method for predicting skin irritation of a prescription containing hydroquinone, which is a first critical stimulus score prediction model that learns paired first critical stimulus learning data to determine a critical stimulus score that causes skin irritation according to hydroquinone application. .

In another aspect of the present invention, the SNPs marker selection device further comprises a step of selecting one or more SNPs related to the presence or absence of skin irritation upon application of hydroquinone from among a plurality of SNPs included in the SNP information, and the irritation score is calculated from a specific individual. It may be determined according to the number and SNPs type of the selected one or more SNPs included in the genetic information.

In another aspect of the present invention, one or more SNPs related to the presence or absence of skin irritation may be any one or more selected from Table 3.

In another aspect of the present invention, different values are assigned to each SNPs type of the major allele - the major allele, the major allele - the minor allele, and the minor allele - the minor allele. The stimulation score may be higher as the number of the selected one or more SNPs increases, and as the number of lower alleles in the selected one or more SNPs increases.

In another aspect of the present invention, the first threshold stimulus score prediction model is generated by the learning processor determining a combination of SNPs used to determine the threshold stimulus score among the selected one or more SNPs, and the stimulus The score can be determined according to the combination of SNPs determined above.

In another aspect of the invention, the threshold stimulation score prediction model is configured to enable the learning processor to send the prescription to an entity having the stimulation score, one or more prescription information from a plurality of prescriptions including a prescription containing hydroquinone, and the SNP information. It may be a second critical stimulation score prediction model that determines the critical stimulation score for each substance according to a plurality of prescriptions by learning second critical stimulation learning data consisting of pairs of stimulation presence or absence when the corresponding substance is applied.

In another aspect of the present invention, the critical stimulus score prediction model includes the learning processor having one or more of the stimulus score, age information, gender information, lifestyle information, environmental information, and skin characteristic information, and the SNP information. A third threshold stimulus score prediction model that determines the threshold stimulus score that causes skin irritation upon application of hydroquinone by learning third threshold stimulus learning data consisting of pairs of stimulus presence and absence when a substance containing hydroquinone is applied to an object. It can be.

In another aspect of the present invention, the critical stimulus score prediction model is such that the learning processor generates a prescription including one or more of the stimulus score, age information, gender information, lifestyle information, environmental information, and skin characteristic information, and hydroquinone. By learning fourth critical stimulus learning data consisting of a pair of prescription information of one or more prescriptions including a plurality of prescriptions and the presence or absence of stimulation when a substance containing hydroquinone is applied to an individual having the SNP information, It may be a fourth critical stimulus score prediction model that determines the critical stimulus score of each substance.

In another aspect of the present invention, (c) the stimulation information calculation unit calculates stimulation information by comparing the stimulation score calculated according to the input SNP information with the threshold stimulation score for each prescription of the plurality of prescriptions, and (d) ) The prescription selection unit may further include a step of selecting one or more prescriptions suitable to be prescribed to an individual with the input SNP information among the plurality of prescriptions using the stimulus information for each prescription.

In another aspect of the present invention, in step (b), the prediction terminal queries the SNP information to a skin brightness change prediction model pre-stored in the prediction terminal or server to calculate predicted skin brightness change information, and calculates It further includes the step of outputting predicted skin brightness change information, wherein the skin brightness change prediction model includes SNP information and skin brightness change information when a learning processor applies a substance containing hydroquinone to an individual having the SNP information. By learning pairs of first skin brightness change learning data, when SNP information is queried, predicted skin brightness change information when a substance containing hydroquinone is applied to an individual with the queried SNP information may be additionally output. there is.

In another aspect of the present invention, the prescription selection unit uses one or more of the presence or absence of stimulation and the predicted skin brightness change information to determine one or more prescriptions suitable to be prescribed to an individual with the queried SNP information among a plurality of prescriptions containing hydroquinone. It may further include a step of selecting.

In addition, one aspect of the present invention is to learn first threshold stimulus learning data consisting of a stimulus score determined according to SNP information and a pair of stimulus presence or absence when a substance containing hydroquinone is applied to an individual having the SNP information to determine the threshold. A server that generates a stimulation score prediction model, which receives the SNP information of the prediction target entity and a server that generates a first threshold stimulation score prediction model that determines the threshold stimulation score that causes skin irritation according to hydroquinone application, and receives the SNP information from the server. Hydroquinone, comprising a prediction terminal that compares the determined threshold stimulation score with the stimulation score determined according to the input SNP information, and outputs stimulation information when hydroquinone is applied to an individual with the input SNP information according to the comparison result. Provides a skin irritation prediction system for prescriptions that include:

In another aspect of the present invention, it further includes a SNPs marker selection device that selects one or more SNPs related to the presence or absence of skin irritation upon application of hydroquinone among the plurality of SNPs included in the SNP information, and the stimulation score is determined from the gene of a specific individual. It may be determined according to the number and SNPs type of the selected one or more SNPs included in the information.

In another aspect of the present invention, a stimulus information calculation unit that calculates a stimulus score according to the input SNP information and calculates the stimulus information by comparing the calculated stimulus score with a threshold stimulus score for each prescription of the plurality of prescriptions, and each of the above It may further include a prescription selection unit that selects one or more prescriptions suitable to be prescribed to an individual with the input SNP information among the plurality of prescriptions using stimulation information for each prescription.

In another aspect of the present invention, using one or more of the stimulation information and the predicted skin brightness change information, one or more prescriptions suitable to be prescribed to an individual with the queried SNP information are selected from among a plurality of prescriptions containing hydroquinone. It may further include a prescription selection unit.

Additionally, one aspect of the present invention provides a computer program stored on a computer-readable recording medium to execute the above-described method.

Another aspect of the present invention provides a system for predicting changes in skin brightness and irritation of a prescription containing hydroquinone. This will be described in detail with reference to FIGS. 1 to 5.

The skin brightness change and/or irritation prediction system includes a prediction terminal 100 and a server 200, and may optionally further include a data providing server 300. Here, the prediction terminal 100 and the server 200 may be separate configurations or may be integrated into one configuration.

When a substance containing hydroquinone is applied to the prediction target object, the prediction terminal 100 predicts predicted skin brightness change information and/or stimulation information in the prediction target object.

Referring to Figure 2, the prediction terminal 100 includes a communication unit 110, an input unit 120, a memory 130, a processor 140, an output unit 150, a control unit 160, a power supply unit 170, and an interface. Includes part 180.

The communication unit 110 is configured for communication with an external device, and data may be transmitted and received with an external device through the communication unit 110.

The input unit 120 may receive commands or data to be used for a component (eg, processor) of the prediction terminal 100 from outside the server 100 (eg, a user). The input unit 120 may include, for example, a microphone, mouse, keyboard, keys (eg, buttons), or a digital pen (eg, stylus pen).

The memory 130 can store various data used in the prediction terminal 100. Data stored in the memory 130 may be, for example, a prediction model generated by the server 200, and may also include input data or output data for software and commands related thereto.

Additionally, the memory 130 may store both information input to the prediction terminal 100 and information generated by the prediction terminal 100.

The processor 140 may perform a data processing or calculation function and, as at least part of the data processing or calculation function, store instructions or data received from other components in a volatile memory, and store the instructions or data stored in the volatile memory. can be processed, and the resulting data can be stored in non-volatile memory. The processor 140 may include a skin brightness change information calculation unit 141, a stimulation information calculation unit 142, and a prescription selection unit 143.

The skin brightness change information calculation unit 141 uses a prediction model generated by the server 200, which will be described later, to calculate predicted skin brightness change information when a substance containing hydroquinone is applied to the prediction target object.

The stimulus information calculation unit 142 uses a prediction model generated by the server 200, which will be described later, to calculate stimulus information upon application of a substance containing hydroquinone to a prediction target object. Here, “stimulus information” is a concept that includes information related to stimulation, such as stimulation grade, presence or absence of stimulation, and stimulation probability, and may be quantified information, but is not particularly limited thereto.

The prescription selection unit 143 uses the information calculated by the skin brightness change information calculation unit 141 and/or the stimulation information calculation unit 142 to calculate and select one or more prescriptions suitable for being prescribed to the corresponding prediction target entity. .

The output unit 150 may output information processed or calculated in the prediction terminal 100 to the outside, and a display and speaker may be included here.

The control unit 160 may control one or more other components (eg, hardware or software components) of the prediction terminal 100 connected to the processor by executing software, for example. The control unit 160 controls the communication unit 110, input unit 120, memory 130, processor 140, output unit 150, power supply unit 170, and interface unit 180 of the prediction terminal 100. can do.

The power supply unit 170 supplies power to the prediction terminal 100. The power supply unit 170 may receive power from an external power source and supply power to the prediction terminal 100.

The interface unit 180 may display information output from the output unit 150 through an interface.

At this time, the interface unit 180 is not limited to a specific type or form.

When a skin brightness change/stimulation prediction command is input through the prediction terminal 100, the processor 140 uses the prediction model generated by the server 200, which will be described later, and stored in the memory 130 of the prediction terminal 100. Thus, predicted skin brightness change information and/or stimulation information is calculated, and the calculated information is output. In another aspect, a skin brightness change/stimulation prediction command may be input through an external input unit, and the input command may be transmitted to the server 200 to perform skin brightness change/stimulation prediction in the server 200.

In another aspect of the present invention, when a skin brightness change/stimulation prediction command is input through the prediction terminal 100, the command is transmitted to the server 200, and using the prediction model stored in the server 200, the server ( Predicted skin brightness change information and/or stimulus information may be calculated at 200, and the prediction terminal 100 may receive the calculated information from the server 200 and output the received information.

Meanwhile, information on the prediction target entity that is the target of the prediction skin brightness change information and/or stimulation information output through the prediction terminal 100 may be input in the form of genetic data (including SNP information), and in another example, the prediction target It may also be entered in the form of entity identification data.

When genetic data of a prediction target entity is input to the prediction terminal 100, the input genetic data may be queried to a prediction model to calculate predicted skin brightness change information and/or stimulation information of the prediction target entity.

On the other hand, when the identification data of the prediction target entity is input to the prediction terminal 100, the input identification data is transmitted to the data providing server 300, and the data providing server 300 determines the entity corresponding to the transmitted identification data. The genetic data is transmitted back to the prediction terminal 100. That is, the genetic data of subjects/prediction target entities is stored in advance in the data provision server 300, and is transmitted when there is a data provision request (genetic data provision request) from the prediction terminal 100 or the server 200, which will be described later. It is possible to transmit genetic data of an individual corresponding to the identified identification data to the prediction terminal 100 or the server 200. In other words, personal information can be protected through a configuration in which genetic data, which is personal information, is stored in a separate data provision server 300 and genetic data of an individual corresponding to the transmitted identification data is provided.

Meanwhile, referring to FIGS. 3 and 6, the server 200 may include a SNPs marker selection device 210, a measurement device 220, a prediction model creation device 230, and a data processing device 240.

The SNPs marker selection device 210 is configured to select one or more SNPs that cause skin brightness change and/or skin irritation when applying a substance containing hydroquinone among a plurality of SNPs. Although it will be described in detail in the following examples, for example, some or all of the 33 SNPs that affect the change in skin brightness by the SNPs marker selection device 210, some or all of the 56 SNPs that affect the degree of skin irritation All SNPs, and some or all SNPs out of 34 that affect the presence or absence of stimulation, may be selected, but are not particularly limited thereto.

The measuring device 220 is configured to obtain information including skin measurement values by measuring the skin of a specific subject. The skin measurement value may include information related to skin condition such as moisture value, skin brightness value, and skin irritation value, but is not limited to the above information as long as it is skin-related information.

Meanwhile, the data processing device 240 is configured to process information measured by the measurement device 220 and data input from the outside. For example, using the subject's skin brightness value before application of a substance containing hydroquinone (L ₀ ) and the subject's skin brightness value (L) after application of a substance containing hydroquinone, the skin brightness change value (ΔL = L - L _O ) can be calculated, and stimulation data can also be calculated by processing the survey results/skin images sent from the subject (H) and the interview results sent from the expert (A) using a predetermined method.

Additionally, the data processing device 240 may improve the performance of the prediction model by adding new information to the information stored in the existing memory.

Information having a predetermined format must be stored in the memory, so that complicated procedures such as additional information conversion and additional preprocessing can be omitted.

The data processing device 240 may perform the role of converting various types of information into a standardized format for each type of information. For example, since information collected from subjects may include personal information, the information can be anonymized, converted into code optimized for computer operations, and functions such as assigning a management number can be performed. In addition, the data processing device 240 can correct errors for each measurement device 220 or standardize measurement information values (for example, the values can be standardized in the manner of (measured value - standard)/standard deviation). This makes it possible to ensure consistency of information.

In addition, the data processing device 240 calculates the outlier value of the information and the outlier value of the information measured by the measurement device 220 by a predetermined method, and if the outlier value is outside the preset value range, the information is processed as noise and stored in the memory. You can prevent it from being stored or output a warning signal so that information can be re-entered/remeasured.

For example, if a standard deviation difference of more than 3 times the average occurs in the menopause response of a male user, the data processing device 240 processes the input information as noise to prevent it from being stored in the memory, or generates a warning signal. It can be output and notified so that information re-entry/re-measurement, etc. can be performed.

The prediction model generating device 230 is configured to generate a model for predicting skin brightness change information/stimulation information when a substance containing hydroquinone is applied.

Referring to FIG. 3, the prediction model generating device 230 may include a communication unit 231, an input unit 232, a processor 233, a memory 234, a learning processor 235, and a control unit 236. The prediction model generating device 230 is also implemented in the form of a computing device capable of calculation, and includes the communication unit 110, input unit 120, memory 130, processor 140, and control unit 160 of the prediction terminal 100 described above. ) A configuration having the same function as ) can be applied.

Meanwhile, the learning processor 235 may perform data processing or calculation functions like the processor 140, but may further include a hardware structure specialized for processing artificial intelligence models.

The memory 234 of the prediction model generating device 230 may include an integrated subject database 234a and a prescription database 234b.

The subject integration database 234a stores information collected from subjects to be used as learning data. The types of information stored in the subject integrated database 234a may include some or all of skin brightness change information (ΔL), genetic information including SNP information, skin irritation information, environmental information, lifestyle information, and skin characteristic information. You can. Here, skin brightness change information (ΔL) refers to the degree of change in skin brightness that occurs when a substance is applied according to a prescription containing hydroquinone alone or optionally additional substances, and skin irritation information refers to the degree of change in skin brightness when the substance is applied. It refers to the degree of skin irritation (or presence or absence of irritation, probability of stimulation) that occurs depending on the condition. Environmental information may include height and weight information, and lifestyle information may include the degree of sunlight exposure, use of UV blockers, eating habits, and smoking status. , may include drinking level, and skin characteristic information may include skin color information, etc., but is not particularly limited thereto.

The prescription database 234b stores information on a number of prescriptions containing hydroquinone. For example, information about a prescription containing only hydroquinone may be stored as Prescription 1, and information about a prescription containing hydroquinone and Salvia Salvia root extract may be stored as Prescription 2. Information may be stored as Prescription 3.

Hereinafter, with reference to FIGS. 4 to 19, the process of generating a prediction model by the prediction model generating device 230 will be described in detail.

The prediction model may be created as the learning processor 235 learns the training data.

In addition, learning here is a concept of deriving an optimal correlation between a plurality of data that is the subject of learning by training a plurality of learning data into a preset artificial neural network (ANN) model. Here, the artificial neural network models include Deep Neural Network (DNN), Convolutional Neural Network (CNN), Deep Convolutional Neural Network (DCNN), Recurrent Neural Network (RNN), Restricted Boltzmann Machine (RBM), Deep Belief Network (DBN), It may be a model based on SSD (Single Shot Detector) or MLP (Multi-layer Perceptron), but is not limited to this and various artificial neural network models can be applied to the present invention. In addition, in another aspect of the present invention, correlations between a plurality of data that are the subject of learning can be derived using various statistical techniques (linear regression model, random forest, lasso, ridge, elastic net, MDR, SVM, etc.) .

In one aspect of the present invention, referring to FIG. 6, the learning processor 235 includes a first pair consisting of SNP information and skin brightness change information (ΔL) when a substance containing hydroquinone is applied to an individual having the SNP information. A first skin brightness change prediction model that learns skin brightness change learning data and, when SNP information is queried, outputs first predicted skin brightness change information when a substance containing hydroquinone is applied to an entity having the queried SNP information. can be created.

In one aspect, the SNP information included in the learning data may include at least some of the one or more SNPs selected by the SNPs marker selection device 210 and may not include SNPs that were not selected. The types of SNPs contained in genes are very diverse. Therefore, if learning is performed considering all SNPs, a problem may occur where learning efficiency is reduced. Accordingly, in the present invention, it may be desirable to adopt a method that improves the efficiency of learning and also improves the accuracy of prediction by learning by considering only the SNPs pre-selected by the SNPs marker selection device 210.

In the process of learning by the learning processor 235, a weight that affects the change in skin brightness when applying a substance containing hydroquinone may be determined for each SNP included in the SNP information.

Here, the weight may be determined by a first value and a second value having different values depending on the SNPs type, and more specifically, it may be determined by multiplying the first value and the second value.

The first value is a value determined according to the number of major alleles and minor alleles included in each SNP. There can be three types of SNPs for each SNP: major allele - major allele, major allele - minor allele, and minor allele - minor allele. The first value is Each type can have different values. More specifically, the first value may have a higher value for a type with a large number of sub-alleles, '1' for major allele-major allele, '2' for major allele-minor allele, and minor allele. In the case of -minor allele, a value of '3' may be assigned, but is not particularly limited thereto.

The second value refers to the influence index by which each SNP affects skin brightness change (more specifically, the influence index by which the minor allele of each SNP affects skin brightness). Since each selected SNP may have a different degree of influence on the change in skin brightness when applying the substance according to the hydroquinone prescription, the present invention calculates the second value, which is the influence index of each SNP, so that arbitrary genetic information can be input later. In this case, predicted skin brightness change information can be calculated using the calculated weight.

Meanwhile, the first value is a value that is fixed and does not change through learning depending on the number of upper alleles and lower alleles included in the SNP, and the second value is a value that changes through learning.

Meanwhile, the first value may be set as a value that changes through learning, and the second value may be fixed to the Effect Size of each SNPs. According to another embodiment, the first value may be fixed according to the number of upper alleles and lower alleles, and the second value may also be fixed as Effect Size.

The learning processor 235 adjusts the weight that affects the skin brightness change for each SNP (specifically, adjusts the second value) to maximize prediction accuracy in the process of learning the first skin brightness change learning data. A first skin brightness change prediction model can be created.

The first skin brightness change prediction model generated by the learning processor 235 may be transmitted to the prediction terminal 100 and stored in the model storage unit 131 of the prediction terminal 100. When genetic data including SNP information of a prediction target entity is queried through the input unit 120 of the prediction terminal 100, the skin brightness change information calculation unit 141 of the prediction terminal 100 uses the first prediction model to 1 Predicted skin brightness change information can be calculated, and the calculated information can be output in the form of visible information to the system user through the output unit 150.

Meanwhile, in another aspect of the present invention, as shown in FIG. 7, identification data of the prediction target entity is input to the prediction terminal 100, the input identification data is transmitted to the data providing server 300, and the data providing server 300 ) may transmit the genetic data of the individual corresponding to the transmitted identification data to the prediction terminal 100.

In one aspect of the present invention, referring to FIG. 8, the learning processor 235 provides SNP information - prescription information of one or more of a plurality of prescriptions including a prescription containing hydroquinone - any one of the prescriptions to an individual having the SNP information. By learning the second skin brightness change learning data consisting of pairs of skin brightness change information when applying a substance according to a prescription, when SNP information is queried, each substance according to a plurality of prescriptions is sent to an individual having the queried SNP information. A second skin brightness change prediction model that outputs second predicted skin brightness change information when applied can be created.

That is, the learning processor 235 adjusts the weight that affects the change in skin brightness for each SNP in each prescription so that the prediction accuracy is maximized in the process of learning the third learning data (specifically, the second value (adjust) to create a second skin brightness change prediction model.

The second skin brightness change prediction model generated by the learning processor 235 may be transmitted to the prediction terminal 100 and stored in the model storage unit 131 of the prediction terminal 100. When genetic data including SNP information of a prediction target entity is queried through the input unit 120 of the prediction terminal 100, the skin brightness change information calculation unit 141 of the prediction terminal 100 creates a second skin brightness change prediction model. Using this method, predicted skin brightness change information for each prescription can be calculated, and the calculated information can be output in the form of visible information to system users through the output unit 150.

Additionally, the prescription selection unit 143 uses the predicted skin brightness change information calculated by the skin brightness change information calculation unit 141 to calculate one or more prescriptions suitable for being prescribed to the corresponding prediction target entity. As an example, a prescription with a large value of the predicted skin brightness change information calculated by the skin brightness change information calculation unit 141 may be calculated as a prescription suitable to be prescribed to the prediction target entity.

Meanwhile, Figure 8 shows the genetic data of the prediction target entity being directly input into the prediction terminal 100. However, as shown in Figure 7, the identification data of the prediction target entity is input into the prediction terminal 100, and the input Identification data is transmitted to the data providing server 300, and the data providing server 300 transmits the genetic data of the individual corresponding to the transmitted identification data to the prediction terminal 100. An aspect may also be included in the scope of the present invention. .

In one aspect of the present invention, referring to FIG. 9, the learning processor 235 is one or more of SNP information - age information / gender information / lifestyle information / environmental information / skin characteristic information - hydroquinone to an individual having the SNP information. By learning the third skin brightness change learning data consisting of a pair of skin brightness change information when a substance containing is applied, one or more of SNP information and age information/gender information/lifestyle information/environment information/skin characteristic information When information is queried, a prediction model that outputs third predicted skin brightness change information when a substance containing hydroquinone is applied to an individual having the queried SNP information can be created.

That is, so that the learning processor 235 maximizes prediction accuracy in the process of learning the third skin brightness change learning data, for each SNP, and one of age information/gender information/lifestyle information/environment information/skin characteristic information A third skin brightness change prediction model can be created by adjusting the weights that affect the skin brightness change.

The third skin brightness change prediction model generated by the learning processor 235 may be transmitted to the prediction terminal 100 and stored in the model storage unit 131 of the prediction terminal 100. When one or more of genetic data including SNP information of the prediction target entity and age information/gender information/lifestyle information/environment information/skin characteristic information are queried through the input unit 120 of the prediction terminal 100, the prediction terminal The skin brightness change information calculation unit 141 of 100 can calculate the third predicted skin brightness change information using the third skin brightness change prediction model, and the calculated information is provided to the system user through the output unit 150. It can be output in the form of visible information.

In one aspect of the present invention, referring to FIG. 10, the learning processor 235 includes SNP information - one or more prescription information among a plurality of prescriptions including a prescription containing hydroquinone - age information/gender information/lifestyle information/environment information /One or more of the skin characteristic information - By learning the fourth skin brightness change learning data consisting of pairs of skin brightness change information when a substance according to one of the prescriptions is applied to an individual having the SNP information, When one or more of SNP information and age information/gender information/lifestyle information/environmental information/skin characteristic information is queried, fourth prediction when each of the substances according to a plurality of prescriptions is applied to an individual with the queried SNP information A prediction model that outputs skin brightness change information can be created.

That is, so that the learning processor 235 maximizes prediction accuracy in the process of learning the fourth skin brightness change learning data, for each SNP in each prescription, and age information/gender information/lifestyle information/environment information/skin A fourth skin brightness change prediction model can be created by adjusting the weights that affect the skin brightness change by adjusting the weights that affect the skin brightness change by one or more of the characteristic information.

The third skin brightness change prediction model generated by the learning processor 235 may be transmitted to the prediction terminal 100 and stored in the model storage unit 131 of the prediction terminal 100. When one or more of genetic data including SNP information of the prediction target entity and age information/gender information/lifestyle information/environment information/skin characteristic information are queried through the input unit 120 of the prediction terminal 100, the prediction terminal The skin brightness change information calculation unit 141 of 100 can calculate predicted skin brightness change information for each prescription using the fourth skin brightness change prediction model, and the calculated information is output through the output unit 150. It can be output in the form of visible information to system users.

Meanwhile, Figure 10 shows the genetic data of the prediction target entity being directly input to the prediction terminal 100. However, as shown in Figure 11, the identification data of the prediction target entity is input into the prediction terminal 100, and the input Identification data is transmitted to the data providing server 300, and the data providing server 300 transmits the genetic data of the individual corresponding to the transmitted identification data to the prediction terminal 100. An aspect may also be included in the scope of the present invention. .

In one aspect of the present invention, referring to FIG. 12, the learning processor 235 is a pair of SNP information - stimulation data (including the presence or absence of stimulation or degree of stimulation) when a substance containing hydroquinone is applied to an entity having the SNP information. By learning the first stimulus learning data consisting of, when SNP information is queried, a prediction model can be created that outputs the first stimulus information when a substance containing hydroquinone is applied to an entity having the queried SNP information.

That is, the learning processor 235 adjusts the weight that affects skin stimulation for each SNP (specifically, adjusts the second value) so that the prediction accuracy is maximized in the process of learning the first stimulus learning data. 1 A stimulus prediction model can be created.

The first stimulus prediction model generated by the learning processor 235 may be transmitted to the prediction terminal 100 and stored in the model storage unit 131 of the prediction terminal 100. When genetic data including SNP information of a prediction target entity is queried through the input unit 120 of the prediction terminal 100, the stimulus information calculation unit 142 of the prediction terminal 100 uses the first stimulus prediction model to generate the first stimulus information. Stimulus information (which may be in the form of the presence or absence of stimulation or the degree of stimulation) can be calculated, and the calculated information can be output in the form of visible information to the system user through the output unit 150.

Meanwhile, Figure 12 shows the genetic data of the prediction target entity being directly input into the prediction terminal 100. However, as shown in Figure 7, the identification data of the prediction target entity is input into the prediction terminal 100, and the input Identification data is transmitted to the data providing server 300, and the data providing server 300 transmits the genetic data of the individual corresponding to the transmitted identification data to the prediction terminal 100. An aspect may also be included in the scope of the present invention. .

In one aspect of the present invention, referring to FIG. 13, the learning processor 235 provides SNP information - prescription information of one or more of a plurality of prescriptions including a prescription containing hydroquinone - any one of the prescriptions to an individual having the SNP information. By learning the second stimulus learning data consisting of pairs of stimulus information when a substance according to a prescription is applied, when SNP information is queried, when each substance according to a plurality of prescriptions is applied to an entity having the queried SNP information A prediction model that outputs the second stimulus information can be created.

That is, the learning processor 235 adjusts the weight that affects skin irritation for each SNP in each prescription so that the prediction accuracy is maximized in the process of learning the second stimulus learning data (specifically, the second value (adjust) to create a second stimulus prediction model.

The second stimulus prediction model generated by the learning processor 235 may be transmitted to the prediction terminal 100 and stored in the model storage unit 131 of the prediction terminal 100. When genetic data including SNP information of a prediction target entity is queried through the input unit 120 of the prediction terminal 100, the stimulus information calculation unit 142 of the prediction terminal 100 uses the second stimulus prediction model to determine each Stimulation information for each prescription can be calculated, and the calculated information can be output in the form of visible information to the system user through the output unit 150.

Additionally, the prescription selection unit 143 uses the stimulus information calculated by the stimulus information calculation unit 142 to calculate one or more prescriptions suitable for being prescribed to the corresponding prediction target entity. As an example, a prescription with a small value of the stimulus information calculated by the stimulus information calculation unit 142 may be calculated as a prescription suitable for being prescribed to the prediction target entity.

Meanwhile, Figure 13 shows the genetic data of the prediction target entity being directly input into the prediction terminal 100. However, as shown in Figure 7, the identification data of the prediction target entity is input into the prediction terminal 100, and the input Identification data is transmitted to the data providing server 300, and the data providing server 300 transmits the genetic data of the individual corresponding to the transmitted identification data to the prediction terminal 100. An aspect may also be included in the scope of the present invention. .

In one aspect of the present invention, referring to FIG. 14, the learning processor 235 is one or more of SNP information - age information / gender information / lifestyle information / environmental information / skin characteristic information - hydroquinone to an individual having the SNP information. By learning the third stimulus learning data consisting of a pair of stimulus information when a substance containing In this case, a prediction model that outputs third stimulus information when a substance containing hydroquinone is applied to an individual with the queried SNP information can be created.

That is, the learning processor 235 provides for each SNP and at least one of age information/gender information/lifestyle information/environmental information/skin characteristic information to maximize prediction accuracy in the process of learning the third stimulus learning data. A third stimulus prediction model can be created by adjusting the weights that affect skin irritation.

The third stimulus prediction model generated by the learning processor 235 may be transmitted to the prediction terminal 100 and stored in the model storage unit 131 of the prediction terminal 100. When one or more of genetic data including SNP information of the prediction target entity and age information/gender information/lifestyle information/environment information/skin characteristic information are queried through the input unit 120 of the prediction terminal 100, the prediction terminal The stimulus information calculation unit 142 of 100 can calculate the third stimulus information using the third stimulus prediction model, and the calculated information is output in the form of information visible to the system user through the output unit 150. It can be.

In one aspect of the present invention, referring to FIG. 15, the learning processor 235 includes SNP information - one or more prescription information among a plurality of prescriptions including a prescription containing hydroquinone - age information/gender information/lifestyle information/environmental information /One or more of the skin characteristic information - learning the fourth stimulus learning data consisting of a pair of stimulus information when a substance according to one of the prescriptions is applied to an individual having the SNP information, SNP information and age When one or more of information/gender information/lifestyle information/environmental information/skin characteristic information is queried, fourth stimulus information is output when each of the substances according to a plurality of prescriptions is applied to an individual having the queried SNP information. A prediction model can be created.

That is, the learning processor 235 adjusts the weight that affects skin irritation for each SNP in each prescription so that the prediction accuracy is maximized in the process of learning the fourth stimulus learning data (specifically, the second value (adjust) can create a fourth stimulus prediction model.

The fourth stimulus prediction model generated by the learning processor 235 may be transmitted to the prediction terminal 100 and stored in the model storage unit 131 of the prediction terminal 100. When genetic data including SNP information of a prediction target entity is queried through the input unit 120 of the prediction terminal 100, the stimulus information calculation unit 142 of the prediction terminal 100 uses the fourth stimulus prediction model to Stimulation information for each prescription can be calculated, and the calculated information can be output in the form of visible information to the system user through the output unit 150.

Meanwhile, Figure 15 shows the genetic data of the prediction target entity being directly input to the prediction terminal 100. However, the identification data of the prediction target entity is input to the prediction terminal 100, and the input identification data is sent to the data providing server. It is transmitted to 300, and the data providing server 300 transmits the genetic data of the individual corresponding to the transmitted identification data to the prediction terminal 100, which may also be included in the scope of the present invention.

In one aspect of the present invention, referring to FIG. 16, the learning processor 235 determines the irritation score according to the SNP information of each individual - the presence or absence of skin irritation when applying a substance containing hydroquinone to an individual having the SNP information. By learning the first critical stimulus learning data consisting of pairs of (i.e., stimulus present for an object that experienced a stimulus and no stimulus for an object that experienced no particular stimulus), when a substance containing hydroquinone was applied, A prediction model that outputs a threshold irritation score that causes skin irritation can be created. Here, skin irritation may include erythema, skin peeling, burning, itching, etc., and is not particularly limited thereto, but is a general term for abnormal reactions.

Here, the stimulation score may be determined according to the number and type of SNPs included in the SNP information and selected by the SNPs selection device 210. More specifically, the stimulation score may be calculated using a first value and a second value, where the first value is determined according to the number of upper alleles and lower alleles included in each SNP, and the second value is is the value at which each SNP affects skin irritation and may be a value predetermined through prior experiments.

In other words, the first value can have a higher value as the number of sub-alleles included in each SNP increases, such as '1' for major allele-major allele, '2' for major allele-minor allele, and '2' for major allele-minor allele. In the case of a minor allele, a value of '3' may be assigned, but is not particularly limited thereto.

The critical stimulus score determined (predicted) in the first critical stimulus score prediction model generated by the learning processor 235 may be transmitted to the prediction terminal 100 and stored in the memory 130 of the prediction terminal 100. . When genetic data including SNP information of a prediction target entity is queried through the input unit 120 of the prediction terminal 100, the stimulation information calculation unit 142 of the prediction terminal 100 calculates a stimulation score according to the queried SNP information. And, stimulation information can be calculated using the comparison result data obtained by comparing the critical stimulation score determined in the first critical stimulation score prediction model (for example, in the case of an object having a stimulation score greater than the critical stimulation score, the stimulation It can be calculated as if there is a stimulus, and for objects with a stimulus score less than the threshold stimulus score, it can be calculated as if there is no stimulus). The calculated stimulus information may be output in the form of visible information to system users through the output unit 150.

Meanwhile, Figure 16 shows the genetic data of the prediction target entity being directly input to the prediction terminal 100. However, the identification data of the prediction target entity is input to the prediction terminal 100, and the input identification data is sent to the data providing server. It is transmitted to 300, and the data providing server 300 transmits the genetic data of the individual corresponding to the transmitted identification data to the prediction terminal 100, which may also be included in the scope of the present invention.

According to one aspect of the present invention, prior to learning the first critical stimulus learning data, the first critical stimulus learning data is grouped based on one or more information selected from gender, age, environmental information, lifestyle information, and skin characteristic information. (For example, when gender is the standard, data obtained from a male individual is grouped into data obtained from a female individual, respectively), and a first threshold stimulus score prediction model can be generated for each grouped data (i.e., generated). A first threshold stimulus score prediction model for each group may be created). In this case, before calculating the stimulus score of the object, the stimulus score is calculated by determining which group the object belongs to according to one or more information selected from gender, age, environmental information, lifestyle information, and skin characteristics, and making the seventh prediction. The presence or absence of stimulation can be predicted through the critical stimulation score determined in the first critical stimulation score prediction model of the corresponding group among the models. Selection of prediction models for each group and prediction of skin irritation according to gender, age, environmental information, lifestyle information, and skin characteristics are performed using the above-mentioned skin brightness change prediction model and irritation prediction model described herein, and the critical irritation score prediction model described later. The same can be applied in .

In one aspect of the present invention, referring to FIG. 17, the learning processor 235 determines a stimulation score according to the SNP information of each individual - one or more prescription information among a plurality of prescriptions including a prescription containing hydroquinone - the SNP information By learning the eighth learning data consisting of pairs of the presence or absence of skin irritation when applying the substance according to each prescription to an object having It is possible to create a prediction model in which the triggered threshold stimulus score is output.

That is, when the learning processor 235 learns the second critical stimulus learning data and generates a second critical stimulus score prediction model, the critical stimulus score for each of the plurality of prescriptions can be determined. The critical stimulus scores determined in the second critical stimulus score prediction model generated by the learning processor 235 may be transmitted to the prediction terminal 100 and stored in the memory 130 of the prediction terminal 100. When genetic data including SNP information of a prediction target entity is queried through the input unit 120 of the prediction terminal 100, the stimulation information calculation unit 142 of the prediction terminal 100 calculates a stimulation score according to the queried SNP information. And, stimulation information for each prescription can be calculated using comparison result data obtained by comparing with the critical stimulation score determined in the second critical stimulation score prediction model (for example, more than the critical stimulation score according to prescription A. In the case of an object with a large irritation score, it is calculated that there is irritation when the substance according to prescription A is applied, and in the case of an object with a stimulation score smaller than the threshold stimulation score according to prescription B, stimulation occurs when the substance according to prescription B is applied. can be calculated without it). The calculated stimulus information may be output in the form of visible information to system users through the output unit 150.

The prescription selection unit 143 may select one or more prescriptions calculated as non-stimulating from among the plurality of prescriptions as appropriate to be prescribed to the subject and output the calculated prescription information through the output unit 150.

Meanwhile, Figure 17 shows the genetic data of the prediction target entity being directly input to the prediction terminal 100. However, the identification data of the prediction target entity is input to the prediction terminal 100, and the input identification data is sent to the data providing server. It is transmitted to 300, and the data providing server 300 transmits the genetic data of the individual corresponding to the transmitted identification data to the prediction terminal 100, which may also be included in the scope of the present invention.

In another aspect of the present invention, referring to FIG. 18, the learning processor 235 may perform a stimulation score - one or more of age information/gender information/lifestyle information/environmental information/skin characteristic information-hydroquinone to an individual having the SNP information. By learning the third critical stimulus learning data consisting of pairs of the presence or absence of skin irritation when a substance containing hydroquinone is applied, a prediction model is generated that outputs a threshold irritation score that causes skin irritation when a substance containing hydroquinone is applied. can do. Here, age information/gender information/lifestyle information/environmental information/skin characteristic information are variables that affect the stimulation score.

The critical stimulus score determined (predicted) in the third critical stimulus score prediction model generated by the learning processor 235 may be transmitted to the prediction terminal 100 and stored in the memory 130 of the prediction terminal 100. . When one or more of genetic data including SNP information of the prediction target object and age information/gender information/lifestyle information/environment information/skin characteristic information are queried through the input unit 120 of the prediction terminal 100, prediction The stimulation information calculation unit 142 of the terminal 100 calculates a stimulation score according to the input SNP information and age information/gender information/lifestyle information/environmental information/skin characteristic information, and calculates the stimulation score determined in the third threshold stimulation score prediction model. Stimulus information can be calculated through comparison result data obtained by comparing with the threshold stimulus score. The calculated stimulus information may be output in the form of visible information to system users through the output unit 150.

Meanwhile, Figure 18 shows the genetic data of the prediction target entity being directly input to the prediction terminal 100. However, the identification data of the prediction target entity is input to the prediction terminal 100, and the input identification data is sent to the data providing server. It is transmitted to 300, and the data providing server 300 transmits the genetic data of the individual corresponding to the transmitted identification data to the prediction terminal 100, which may also be included in the scope of the present invention.

In one aspect of the present invention, referring to FIG. 19, the learning processor 235 determines a stimulation score - one or more prescription information among a plurality of prescriptions including a prescription containing hydroquinone - age information/gender information/lifestyle information/environmental information. /One or more of the skin characteristic information - By learning the 9th learning data consisting of pairs of skin irritation when applying the substance according to each prescription to an individual having the SNP information, each substance according to a plurality of prescriptions is learned. A prediction model can be created that outputs a threshold irritation score that causes skin irritation for each of multiple prescriptions when applied. Here, age information/gender information/lifestyle information/environmental information/skin characteristic information are variables that affect the stimulation score.

The critical stimulus scores determined (predicted) in the fourth critical stimulus score prediction model generated by the learning processor 235 may be transmitted to the prediction terminal 100 and stored in the memory 130 of the prediction terminal 100. . When one or more of genetic data including SNP information of the prediction target entity and age information/gender information/lifestyle information/environment information/skin characteristic information are queried through the input unit 120 of the prediction terminal 100, the prediction terminal The stimulation information calculation unit 142 of 100 calculates a stimulation score according to the queried SNP information/age information/gender information/lifestyle information/environmental information/skin characteristic information, and the threshold determined in the fourth threshold stimulation score prediction model. Stimulation information for each prescription can be calculated using the comparison result data obtained by comparing the stimulation score (for example, in the case of an object having a stimulation score greater than the threshold stimulation score according to A prescription, the stimulation information according to A prescription can be calculated. It is calculated that there is irritation when the substance is applied, and in the case of an object with an irritation score that is less than the threshold irritation score according to prescription B, it can be calculated that there is no irritation when the substance according to prescription B is applied). The calculated stimulus information may be output in the form of visible information to system users through the output unit 150.

Meanwhile, Figure 19 shows the genetic data of the prediction target entity being directly input to the prediction terminal 100. However, the identification data of the prediction target entity is input to the prediction terminal 100, and the input identification data is sent to the data providing server. It is transmitted to 300, and the data providing server 300 transmits the genetic data of the individual corresponding to the transmitted identification data to the prediction terminal 100, which may also be included in the scope of the present invention.

The prediction terminal 100 according to an aspect of the present invention includes at least one skin among the above-described first to fourth skin brightness change prediction models, first to fourth stimulus prediction models, and first to fourth critical stimulus score prediction models. An operation for predicting skin brightness change by a brightness change prediction model, and an operation for predicting skin irritation information by at least one stimulus prediction model, or a critical stimulus score determined in at least one threshold stimulus score prediction model and genetic information. Calculations for predicting skin irritation information through comparison of possible irritation scores can be performed simultaneously, and output according to the calculations can also be performed.

In this way, in addition to the predicted skin brightness change information output through the calculation of the skin brightness change prediction model, the predicted stimulus information output through the calculation of the stimulus prediction model and the critical stimulus score and genetic information determined in the threshold stimulus score prediction model The following information can be further provided to the user by using one or more of the stimulus information output through a comparison operation of the stimulus score that can be determined.

In other words, the prediction terminal 100 may further calculate whether the prediction target object for which a query is made belongs to one of the four groups through a combination of the predicted skin brightness change information and the predicted stimulus information.

The above four groups are as follows.

First, it is a suitable group. When the value of the predicted skin brightness change information is greater than the first reference value and the value of the predicted stimulus information is smaller than the second reference value, the prediction terminal 100 calculates the prediction target object as a “suitable group” and prescribes the prescription selection unit ( 143) can calculate a prescription for hydroquinone alone and provide it to the user.

Second, there is a high stimulation group. When the value of the predicted skin brightness change information is greater than the first reference value and the value of the predicted stimulus information is greater than the second reference value, the prediction terminal 100 calculates the prediction target object as a “high stimulation group” and prescribes the prescription selection unit. (143) can calculate a prescription in which an irritation-relieving substance is added in addition to hydroquinone and provide it to the user.

Third, there is a low-efficacy group. When the value of the predicted skin brightness change information is smaller than the first reference value and the value of the predicted stimulus information is smaller than the second reference value, the prediction terminal 100 calculates the prediction target entity as a “low-efficacy group” and prescribes the prescription selection unit. (143) can calculate and provide a prescription to the user in which an efficacy synergistic substance that has an effect of improving skin brightness is added in addition to hydroquinone.

Fourth, it is a non-suitable group. When the value of the predicted skin brightness change information is smaller than the first reference value and the value of the predicted stimulus information is larger than the second reference value, the prediction terminal 100 calculates the prediction target object as a “non-suitable group” and prescribes the prescription selection unit. (143) can calculate and provide a prescription to the user in which efficacy synergistic substances and irritation relieving substances are added in addition to hydroquinone.

Meanwhile, the form of output provided to the user may be in the form of Figure 39 or Figure 40, but is not particularly limited thereto.

Hereinafter, a method for predicting changes in skin brightness and irritation of a prescription containing hydroquinone according to another aspect of the present invention will be described in detail with reference to FIGS. 20 to 34. The method for predicting skin brightness change and irritation according to an aspect of the present invention can be performed using the prediction model generated by the server 200 described above.

Referring to FIG. 20, subject phenotype data (skin brightness change information, stimulation information, stimulation score) and genotype data (SNP information) are acquired, and the SNPs marker selection device 210 uses the acquired data to select the skin when hydroquinone is applied. Select one or more SNPs markers that affect brightness changes and/or skin irritation. Next, the learning processor 235 of the server 200 uses learning data (first to fourth skin brightness change learning data, first to fourth stimulus learning data, and first to fourth stimulation learning data) to derive a correlation between the phenotype data and the genotype data. 1 to 4 threshold stimulus learning data) to learn any one of the 1st to 4th skin brightness change prediction model, the 1st to 4th stimulus prediction model, and the 1st to 4th threshold stimulus score prediction model. Create. Next, the genetic data or identification data of the prediction target entity is input, the genetic data is queried to the generated prediction model, and one or more of predicted skin brightness change information, stimulation information, and threshold stimulation score is calculated in the prediction model. Next, using the calculated information, the prescription selection unit 143 calculates one or more prescriptions suitable to be prescribed to the corresponding prediction target entity.

Referring to FIG. 21, the process of selecting one or more SNPs markers that affect the change in skin brightness when hydroquinone is applied by the SNPs marker selection device 210 will be described in detail.

First, the skin of the subject H may be measured by the measuring device 220 before application of the product containing hydroquinone, and the skin brightness value (L ₀ ) of the subject H may be calculated by the data processing device 240.

Next, after applying the hydroquinone-containing product to the subject H, the skin of the subject H is measured by the measuring device 220, and the skin brightness value (L) of the subject H is measured by the data processing device 240. ) can be calculated.

Next, the data processing device 240 calculates the skin brightness change value (ΔL = L - L ₀ ) when the hydroquinone-containing material is applied to the subject, and calculates the calculated skin brightness change value to the SNPs marker selection device 210. send to

Meanwhile, the SNPs marker selection device 210 performs correlation analysis between the subject's genetic data (which may be stored on a server or may be obtained by transmitting subject identification information to a data providing server) and the skin brightness change value. , one or more skin brightness change prediction SNPs markers that affect skin brightness change when applying a substance containing hydroquinone can be selected. The skin brightness change prediction SNPs markers selected in this way can be transmitted to the prediction model creation device 230 and can be used as an element of learning data to create a prediction model in the future.

Referring to FIG. 22, the process of selecting one or more SNPs markers that affect skin irritation when applying hydroquinone by the SNPs marker selection device 210 will be described in detail.

After the hydroquinone-containing product is applied to the subject (H), the subject can be interviewed by an expert (A), or the subject (H) can take a survey/skin image on its own. The interview results and/or survey results/skin images are transmitted to the data processing device 240, and the data processing device 240 can calculate stimulation data (degree of stimulation and/or presence or absence of stimulation) using the transmitted data. there is.

Meanwhile, the SNPs marker selection device 210 performs correlation analysis between the subject's genetic data (which may be stored on a server or may be obtained by transmitting subject identification information to a data providing server) and the skin brightness change value. , one or more irritation prediction SNPs markers that affect skin irritation when applying a substance containing hydroquinone can be selected. The stimulus prediction SNPs markers selected in this way can be transmitted to the prediction model creation device 230 and can be used as an element of learning data to create a prediction model in the future.

Referring to FIGS. 23 to 30, the prediction model generating device 230 of the server 200 learns the first to fourth skin brightness change learning data and the first to fourth stimulus data to generate the first to fourth skin brightness change. A prediction model and first to fourth stimulus prediction models are generated, and the generated prediction model is transmitted to the prediction terminal 100.

A skin brightness change and/or stimulation prediction command is input through the prediction terminal 100, and genetic data (can be input directly or transmitted through a data providing server) and/or age/ Using one or more data of gender/lifestyle/environment/skin characteristic information, skin brightness change information and/or stimulation information when applying a substance containing hydroquinone to the prediction target object may be calculated.

This information can be calculated for each of a plurality of prescriptions, and the prescription selection unit selects and outputs one or more prescriptions suitable to be prescribed to the prediction target entity using the calculated skin brightness change information and/or stimulation information.

Referring to FIGS. 31 to 34, the prediction model generating device 230 of the server 200 learns the first to fourth critical stimulus learning data to determine the critical stimulus score that causes skin irritation when applying a substance containing hydroquinone. The first to fourth critical stimulus score prediction models are generated, and the determined critical stimulus score is transmitted to the prediction terminal 100.

A stimulus prediction command is input through the prediction terminal 100, and genetic data (can be input directly or transmitted through a data providing server) and/or age/gender/lifestyle/environment of the prediction target object that is the subject of prediction /Using one or more data from the skin characteristic information, the stimulation score of the prediction target object may be calculated.

The stimulus information calculation unit compares the calculated stimulus score and the determined threshold stimulus score, and calculates stimulus information using the comparison result data.

This information can be calculated for each of a plurality of prescriptions, and the prescription selection unit uses the calculated stimulus information to select and output one or more prescriptions suitable to be prescribed to the corresponding prediction target entity.

The method according to an embodiment of the present invention described above may be implemented in the form of computer program instructions that can be executed through various computer means and recorded on a computer-readable medium. The computer-readable medium may include program instructions, data files, data structures, etc., singly or in combination. Program instructions recorded on the medium may be specially designed and constructed for the present invention or may be known and usable by those skilled in the art of computer software. Examples of computer-readable recording media include magnetic media such as hard disks, floppy disks, and magnetic tapes, optical media such as CD-ROMs and DVDs, and magnetic media such as floptical disks. -Includes optical media (magneto-optical media) and hardware devices specifically configured to store and execute program instructions, such as ROM, RAM, flash memory, etc. Examples of program instructions include machine language code, such as that produced by a compiler, as well as high-level language code that can be executed by a computer using an interpreter, etc. The hardware devices described above may be configured to operate as one or more software modules to carry out the operations of the present invention, and vice versa.

Through the prediction method and system according to aspects of the present invention, it is possible to provide information on prediction of skin brightness change and/or stimulation prediction upon application of a substance according to a plurality of prescriptions containing hydroquinone.

Additionally, based on predicted skin brightness change information and/or stimulation information, it is possible to select and provide a prescription suitable to be prescribed to the user among a plurality of prescriptions.

In addition, it is possible to provide the most effective prescription to an individual based on information on genetic polymorphism markers observed in the individual, while providing the optimal prescription that further takes into account the individual's environmental information/lifestyle information/skin characteristics information/age information/gender information. It is possible to select and provide.

In addition, in the process of predicting changes in skin brightness and/or predicting stimulation information, the accuracy of prediction is high because the results of machine learning are used.

In addition, it is possible to provide information to select an irritant reliever that can relieve irritation caused by hydroquinone according to information on genetic polymorphism markers observed in an individual.

1 is a schematic block diagram illustrating one aspect of a system for predicting changes in skin brightness and/or skin irritation of a prescription containing hydroquinone according to the present invention.

FIG. 2 is a schematic block diagram for explaining an aspect of a prediction terminal of the system of FIG. 1.

FIG. 3 is a schematic block diagram illustrating an aspect of a prediction model generating apparatus of the system of FIG. 1 .

Figure 4 is a diagram for explaining data transmitted and received between the server of the system of Figure 1, the prediction terminal, and the data providing server.

FIG. 5 is a diagram illustrating how information is calculated in a prediction model according to data input to a prediction terminal and the calculated information is output in one aspect of the present invention.

Figure 6 illustrates that in one aspect of the present invention, a first skin brightness change prediction model is generated through learning, and information calculated from the first skin brightness change prediction model is output according to data input to the prediction terminal. This is a drawing for this purpose.

FIG. 7 is a diagram illustrating an aspect different from FIG. 6 in which identification data of a prediction target entity is transmitted to a prediction terminal, genetic data corresponding to the identification data is transmitted from a data providing server, and calculation is performed.

Figure 8 illustrates that in one aspect of the present invention, a second skin brightness change prediction model is generated through learning, and information calculated from the second skin brightness change prediction model is output according to data input to the prediction terminal. This is a drawing for this purpose.

Figure 9 illustrates that in one aspect of the present invention, a third skin brightness change prediction model is generated through learning, and information calculated from the third skin brightness change prediction model is output according to data input to the prediction terminal. This is a drawing for this purpose.

Figure 10 illustrates that in one aspect of the present invention, a fourth skin brightness change prediction model is generated through learning, and information calculated from the fourth skin brightness change prediction model is output according to data input to the prediction terminal. This is a drawing for this purpose.

FIG. 11 is a diagram for explaining an aspect different from FIG. 10 in which identification data of a prediction target entity is transmitted to a prediction terminal, genetic data corresponding to the identification data is transmitted from a data providing server, and calculation is performed.

FIG. 12 is a diagram illustrating how, in one aspect of the present invention, a first stimulus prediction model is generated through learning and information calculated from the first stimulus prediction model is output according to data input to a prediction terminal. .

FIG. 13 is a diagram illustrating how, in one aspect of the present invention, a second stimulus prediction model is generated through learning and information calculated from the second stimulus prediction model is output according to data input to a prediction terminal. .

FIG. 14 is a diagram illustrating how, in one aspect of the present invention, a third stimulus prediction model is generated through learning and stimulus information calculated according to data input to a prediction terminal is output.

FIG. 15 is a diagram illustrating how, in one aspect of the present invention, a fourth stimulus prediction model is generated through learning and stimulus information calculated according to data input to a prediction terminal is output.

FIG. 16 is a diagram illustrating how, in one aspect of the present invention, a first threshold stimulus score prediction model is generated through learning and stimulus information calculated according to data input to a prediction terminal is output.

FIG. 17 is a diagram illustrating how, in one aspect of the present invention, a second critical stimulus score prediction model is generated through learning and stimulus information calculated according to data input to a prediction terminal is output.

FIG. 18 is a diagram illustrating how, in one aspect of the present invention, a third threshold stimulus score prediction model is generated through learning and stimulus information calculated according to data input to a prediction terminal is output.

FIG. 19 is a diagram illustrating how, in one aspect of the present invention, a fourth critical stimulus score prediction model is generated through learning and stimulus information calculated according to data input to a prediction terminal is output.

Figure 20 is a schematic flowchart for explaining one aspect of a method for predicting skin brightness change and/or skin irritation of a prescription containing hydroquinone.

Figure 21 is a schematic flowchart illustrating an aspect of the present invention for selecting SNPs that affect changes in skin brightness upon application of a substance containing hydroquinone.

Figure 22 is a schematic flowchart for explaining an aspect of the present invention for selecting SNPs that affect skin irritation when applying a substance containing hydroquinone.

23 to 30 are schematic flowcharts to explain how predicted skin brightness change information, stimulation information, and/or selection prescription information are output as the genetic data of the prediction target entity is queried to the generated prediction model.

Figures 31 to 34 are schematic flowcharts for explaining how stimulus information is calculated by comparing the stimulus score determined according to the genetic data of the prediction target entity and the critical stimulus score determined in each prediction model.

Figure 35 is a graph comparing skin brightness change information predicted based on a prediction model generated in one aspect of the present invention and actually measured skin brightness change information.

Figure 36 is a graph comparing the degree of stimulation predicted based on the prediction model generated in one aspect of the present invention and the degree of stimulation actually measured.

Figure 37 is a graph comparing the presence or absence of stimulation predicted based on a prediction model generated in one aspect of the present invention and the presence or absence of stimulation actually measured.

Figure 38 shows whether hydroquinone irritation is reduced after using the irritation relieving gel according to the present invention.

Figures 39 and 40 are diagrams to explain various output forms that can be output from the prediction terminal as SNP information of the prediction target entity is input.

Hereinafter, the present invention will be described in more detail through examples. These examples are merely for illustrating the present invention, and the scope of the present invention is not to be construed as limited by these examples.

실시예 1: 히드로퀴논 효능과 자극 평가 및 유전자 채집Example 1: Hydroquinone efficacy and stimulation evaluation and gene collection

1-1. 히드로퀴논 효능 평가 및 유전자 채집1-1. Hydroquinone efficacy evaluation and gene collection

In order to derive genetic polymorphism markers that can explain the degree of hydroquinone efficacy (degree of change in skin brightness), 33 healthy Korean subjects in their 20s to 50s were recruited. Localized areas with individual pigmentation were selected and one type of hydroquinone 4% product was applied topically once a day (in the evening) for 4 weeks.

After measuring the ΔL value of the area where the hydroquinone 4% product was applied topically and the area on the other side of the face (ΔL = L value at week 4 - L value at week 0), ΔL of the local area where hydroquinone 4% product was applied - ΔL of the area opposite the face was calculated to determine hydroquinone efficacy (skin) The degree of change in brightness was evaluated.

For skin measurement, all analyzed subjects washed their faces using cleansing products, waited for 30 minutes without applying any products to allow the skin to adapt to the measurement environment, and then evaluated the degree of hydroquinone efficacy (change in skin brightness). For skin evaluation, a Chroma meter CR-400 (KONICA MINOLTA, Japan), a specialized device for measuring skin color, was used, and the average L value obtained by averaging 5 measurements was used.

1-2. 히드로퀴논 자극 평가 및 유전자 채집1-2. Hydroquinone stimulation assessment and gene collection

In order to derive genetic polymorphism markers that can explain the degree/presence of hydroquinone stimulation, 93 healthy Koreans in their 20s to 50s were recruited. After applying one type of hydroquinone 4% product once a day (evening) to localized areas of the face with individual pigmentation for 3 weeks, a questionnaire was administered regarding irritation including itching, stinging, erythema, and rash, and a score of 4 was obtained through face-to-face consultation with an evaluator. The evaluation progressed and the presence or absence of stimulation was determined using the scaling method (0: no stimulation at all, 1: minimal stimulation, 2: weak, 3: strong).

Gene collection was conducted through saliva collection, and for effective gene collection, all analysis subjects were prohibited from consuming any food, including water, from 30 minutes before collection.

Among the above subjects: ① are pregnant, lactating, or planning to become pregnant within 6 months; ② have used dermatological agents containing steroids for more than 1 month to treat skin diseases; ③ no more than 6 months have passed since participating in the same test ④ If you have sensitive or hypersensitive skin; ⑤ If you have skin abnormalities such as moles, acne, erythema, or telangiectasia on the test site; ⑥ If you use the same or similar cosmetics or medicines on the test site within 3 months of starting the test ⑦ If you have received a procedure (skin peeling, Botox, other skin care) on the test area or plan to do so within 6 months, ⑧ If you have a chronic wasting disease (asthma, diabetes, high blood pressure, etc.), ⑨ If you have atopic dermatitis, ⑩ Other cases in which the main investigator judged the test to be difficult were excluded from the list of subjects.

실시예 2: 유전자형 분석Example 2: Genotyping

For gene extraction from saliva for genetic analysis, human genomic DNA was extracted using the QIAamp mini prep kit (QIAGEN), and its quality was determined by absorbance (OD 260/280) or 1.7, concentration 50ng/ul, 1x TAE 1% agarose. It was confirmed through a band test using gel, and genetic analysis was performed only on those that passed the quality test.

Genetic analysis was conducted using an Illumina microarray genotyping chip, and specifically, the genes of the subjects were analyzed using a global screening array product from the same company.

The Illumina microarray genotyping chip genetic analysis experiment was conducted according to the provided manual, and the provided reagents were used to perform genomic DNA amplification, DNA fragmentation, precipitation, hybridization, and staining. , washing, coating, and scanning were performed.

The microarray genotyping chip on which the experiment was completed was scanned using iScan Control Software (Illumina), and when scanning was completed, an idat file was automatically created and data quality was controlled using the GenomeStudio (Illumina) program (sample call rate 98%, marker call). rate 98%) and genetic information confirmation was performed.

In this experiment, only data that passed data quality control after genetic analysis were used.

실시예 3: 히드로퀴논 효능 및 자극 연관 유의성 유전자 다형성 마커 도출Example 3: Derivation of significant genetic polymorphism markers associated with hydroquinone efficacy and stimulation

In order to derive genetic polymorphism markers that have correlation with hydroquinone efficacy (change in skin brightness) and stimulation, linear regression analysis was performed using genetic polymorphism markers of the analysis subjects, and the R v 3.6.1 program was used for analysis.

Imputation analysis was performed using the Beagle v 5.1 program to secure information on additional genetic polymorphism markers using the genetic polymorphism markers identified through experiments. The reference data used to conduct the analysis was the international public genome. Information registered in the 1000 Genomes project, a data database, was used.

Imputation is a statistical technique that infers unanalyzed genetic polymorphism marker information based on genetic polymorphism marker information obtained through experiments. To ensure statistical significance of the genetic polymorphism markers subject to analysis, the analysis was limited to genetic markers involved in skin physiological activities such as skin composition production, melanin biosynthesis, oxidative metabolism, nerve stimulation, and inflammatory response.

The significance of genetic polymorphism markers showing correlation with hydroquinone efficacy and stimulation was evaluated through linear regression analysis F-statistics, and the standard was set at P-value < 0.1.

In order to minimize external effects that may affect hydroquinone efficacy and stimulation and derive effects based on genetic information, information on age, gender, BMI, or lifestyle habits (drinking, smoking, eating habits, sleeping habits, etc.) The stimulus can be corrected and used for analysis (for example, performing linear regression analysis).

A number of significant genetic polymorphism markers associated with hydroquinone efficacy and stimulation were identified (Tables 1 to 3).

Hydroquinone skin brightening improvement efficacy prediction genetic polymorphism marker (P < 0.1)

	표현형phenotype	SNP¹⁾ SNP ¹⁾	Gene¹⁾ Gene ¹⁾	Chr:Position¹⁾ Chr:Position ¹⁾	Allele²⁾ Allele ²⁾	MAF³⁾ MAF ³⁾	P-value⁴⁾ P-value ⁴⁾	Effect size (β)⁵⁾ Effect size (β) ⁵⁾
1One	피부 밝기개선Improved skin brightness	rs7718428rs7718428	PPARGC1BPPARGC1B	5:1491670735:149167073	A>GA>G	0.02442530.0244253	0.00204940.0020494	-2.564856474-2.564856474
22	피부 밝기개선Improved skin brightness	rs12788260rs12788260	TYRT.Y.R.	11:8897696311:88976963	T>CT>C	0.04166670.0416667	0.00393070.0039307	4.066277384.06627738
33	피부 밝기개선Improved skin brightness	rs4396293rs4396293	TYRT.Y.R.	11:8897611311:88976113	T>CT>C	0.29454020.2945402	0.00684050.0068405	-0.991478997-0.991478997
44	피부 밝기개선Improved skin brightness	rs200806138rs200806138	MGST2MGST2	4:1406502724:140650272	A>CA>C	0.02155170.0215517	0.00969730.0096973	-3.690048689-3.690048689
55	피부 밝기개선Improved skin brightness	rs7162117rs7162117	OCA2OCA2	15:2830434415:28304344	A>GA>G	0.01867820.0186782	0.01235060.0123506	-2.65508713-2.65508713
66	피부 밝기개선Improved skin brightness	rs72714114rs72714114	OCA2OCA2	15:2832781915:28327819	A>GA>G	0.01867820.0186782	0.01235060.0123506	-2.65508713-2.65508713
77	피부 밝기개선Improved skin brightness	rs4855447rs4855447	MITFMITF	3:698033283:69803328	A>GA>G	0.42959770.4295977	0.01915720.0191572	-1.05029909-1.05029909
88	피부 밝기개선Improved skin brightness	rs7651218rs7651218	MITFMITF	3:698372833:69837283	T>CT>C	0.42528740.4252874	0.01915720.0191572	-1.05029909-1.05029909
99	피부 밝기개선Improved skin brightness	rs2122005rs2122005	OCA2OCA2	15:2826702715:28267027	A>GA>G	0.04166670.0416667	0.03778630.0377863	1.8344086951.834408695
1010	피부 밝기개선Improved skin brightness	rs73838671rs73838671	MITFMITF	3:699132393:69913239	T>CT>C	0.08333330.0833333	0.04661250.0466125	-1.175879455-1.175879455
1111	피부 밝기개선Improved skin brightness	rs78955679rs78955679	GSTM5GSTM5	1:1102578141:110257814	C>GC>G	0.1250.125	0.05094440.0509444	1.5146704551.514670455
1212	피부 밝기개선Improved skin brightness	rs2594934rs2594934	OCA2OCA2	15:2818271515:28182715	T>CT>C	0.21839080.2183908	0.06161730.0616173	-0.789333093-0.789333093
1313	피부 밝기개선Improved skin brightness	rs8034072rs8034072	OCA2OCA2	15:2818665815:28186658	T>CT>C	0.22126440.2212644	0.06161730.0616173	-0.789333093-0.789333093
1414	피부 밝기개선Improved skin brightness	rs2160512rs2160512	MGST1MGST1	12:1651902012:16519020	A>GA>G	0.04741380.0474138	0.06233360.0623336	-1.208724403-1.208724403
1515	피부 밝기개선Improved skin brightness	rs11970311rs11970311	GSTA1GSTA1	6:526664486:52666448	C>GC>G	0.16522990.1652299	0.06352440.0635244	1.2349653771.234965377
1616	피부 밝기개선Improved skin brightness	rs192182485rs192182485	GSTA5GSTA5	6:527056846:52705684	T>GT>G	0.01005750.0100575	0.06758920.0675892	-1.946106146-1.946106146
1717	피부 밝기개선Improved skin brightness	rs186350312rs186350312	MITFMITF	3:699103053:69910305	A>GA>G	0.0129310.012931	0.0701350.070135	-1.610423779-1.610423779
1818	피부 밝기개선Improved skin brightness	rs149802322rs149802322	KITLGKITLG	12:8890576612:88905766	A>GA>G	0.01580460.0158046	0.07461190.0746119	-3.16660692-3.16660692
1919	피부 밝기개선Improved skin brightness	rs1571858rs1571858	GSTM3GSTM3	1:1102799141:110279914	T>CT>C	0.27442530.2744253	0.08097780.0809778	-0.67523196-0.67523196
2020	피부 밝기개선Improved skin brightness	rs10510992rs10510992	MITFMITF	3:699481683:69948168	T>CT>C	0.08477010.0847701	0.08139210.0813921	-1.036009587-1.036009587
2121	피부 밝기개선Improved skin brightness	rs202171165rs202171165	GSTM3GSTM3	1:1102829091:110282909	T>CT>C	0.00574710.0057471	0.08394190.0839419	2.5440711152.544071115
2222	피부 밝기개선Improved skin brightness	rs182546041rs182546041	PPARGC1BPPARGC1B	5:1491256955:149125695	A>GA>G	0.02298850.0229885	0.08394190.0839419	2.5440711152.544071115
2323	피부 밝기개선Improved skin brightness	rs117925628rs117925628	RAB27ARAB27A	15:5550184415:55501844	A>GA>G	0.02442530.0244253	0.08394190.0839419	2.5440711152.544071115
2424	피부 밝기개선Improved skin brightness	rs4715352rs4715352	GSTA5GSTA5	6:527033966:52703396	T>CT>C	0.21982760.2198276	0.08826640.0882664	-0.810173842-0.810173842
2525	피부 밝기개선Improved skin brightness	rs11969892rs11969892	GSTA2GSTA2	6:526238876:52623887	T>AT>A	0.16091950.1609195	0.09137680.0913768	1.0686816231.068681623
2626	피부 밝기개선Improved skin brightness	rs3957356rs3957356	GSTA1GSTA1	6:526686706:52668670	T>CT>C	0.16235630.1623563	0.09222680.0922268	1.041875891.04187589
2727	피부 밝기개선Improved skin brightness	rs3957357rs3957357	GSTA1GSTA1	6:526686876:52668687	A>GA>G	0.15660920.1566092	0.09222680.0922268	1.041875891.04187589
2828	피부 밝기개선Improved skin brightness	rs75583983rs75583983	GSTA2GSTA2	6:526171796:52617179	T>CT>C	0.15517240.1551724	0.09222680.0922268	1.041875891.04187589
2929	피부 밝기개선Improved skin brightness	rs2608629rs2608629	GSTA2GSTA2	6:526257946:52625794	A>GA>G	0.15517240.1551724	0.09222680.0922268	1.041875891.04187589
3030	피부 밝기개선Improved skin brightness	rs2749008rs2749008	GSTA2GSTA2	6:526261206:52626120	T>CT>C	0.16235630.1623563	0.09222680.0922268	1.041875891.04187589
3131	피부 밝기개선Improved skin brightness	rs7745053rs7745053	GSTA5GSTA5	6:526983406:52698340	A>GA>G	0.15373560.1537356	0.09222680.0922268	1.041875891.04187589
3232	피부 밝기개선Improved skin brightness	rs76801519rs76801519	GSTA5GSTA5	6:526996466:52699646	T>CT>C	0.15373560.1537356	0.09222680.0922268	1.041875891.04187589
3333	피부 밝기개선Improved skin brightness	rs73854327rs73854327	MGST2MGST2	4:1406019824:140601982	A>GA>G	0.07183910.0718391	0.09356030.0935603	-1.768488515-1.768488515

Genetic polymorphism markers predicting degree of hydroquinone stimulation (P < 0.1)

	표현형phenotype	SNP¹⁾ SNP ¹⁾	Gene¹⁾ Gene ¹⁾	Chr:Position¹⁾ Chr:Position ¹⁾	Allele²⁾ Allele ²⁾	MAF³⁾ MAF ³⁾	P-value⁴⁾ P-value ⁴⁾	Effect size (β)⁵⁾ Effect size (β) ⁵⁾
1One	자극정도degree of stimulation	rs12712969rs12712969	PRKCEPRKCE	2:463321692:46332169	T>CT>C	0.03017240.0301724	0.00350110.0035011	1.37455851.3745585
22	자극정도degree of stimulation	rs6720975rs6720975	PRKCEPRKCE	2:460242122:46024212	T>CT>C	0.2629310.262931	0.00465570.0046557	-0.50983-0.50983
33	자극정도degree of stimulation	rs1375055rs1375055	PRKCEPRKCE	2:463403782:46340378	A>GA>G	0.0459770.045977	0.00873740.0087374	1.15883521.1588352
44	자극정도degree of stimulation	rs28902234rs28902234	TRPM8TRPM8	2:2349168342:234916834	A>GA>G	0.11494250.1149425	0.01350440.0135044	1.18754981.1875498
55	자극정도degree of stimulation	rs72808770rs72808770	PRKCEPRKCE	2:463613222:46361322	A>GA>G	0.01724140.0172414	0.01419480.0141948	1.42383491.4238349
66	자극정도degree of stimulation	rs34597632rs34597632	TGM1TGM1	14:2472806114:24728061	A>GA>G	0.08333330.0833333	0.01581310.0158131	0.78669890.7866989
77	자극정도degree of stimulation	rs6752001rs6752001	TRPM8TRPM8	2:2349060462:234906046	T>GT>G	0.11350570.1135057	0.01678080.0167808	-0.577248-0.577248
88	자극정도degree of stimulation	rs3742506rs3742506	TGM1TGM1	14:2472514114:24725141	A>GA>G	0.07471260.0747126	0.03171140.0317114	0.65013060.6501306
99	자극정도degree of stimulation	rs75042186rs75042186	PRKCEPRKCE	2:463670492:46367049	A>CA>C	0.03591950.0359195	0.03320780.0332078	1.01300181.0130018
1010	자극정도degree of stimulation	rs10166692rs10166692	PRKCEPRKCE	2:460770382:46077038	T>CT>C	0.23994250.2399425	0.03693620.0369362	0.38853740.3885374
1111	자극정도degree of stimulation	rs112232129rs112232129	PRKCEPRKCE	2:463241032:46324103	A>GA>G	0.01005750.0100575	0.03934180.0393418	1.66925651.6692565
1212	자극정도degree of stimulation	rs67179108rs67179108	PRKCEPRKCE	2:463250092:46325009	T>GT>G	0.01005750.0100575	0.03934180.0393418	1.66925651.6692565
1313	자극정도degree of stimulation	rs10954810rs10954810	NRG1NRG1	8:315557438:31555743	T>CT>C	0.00718390.0071839	0.04107710.0410771	1.37166031.3716603
1414	자극정도degree of stimulation	rs11989919rs11989919	NRG1NRG1	8:325026268:32502626	T>CT>C	0.46408050.4640805	0.0428710.042871	-0.318473-0.318473
1515	자극정도degree of stimulation	rs7357509rs7357509	SDR16C5SDR16C5	8:572327878:57232787	T>CT>C	0.04454020.0445402	0.04297780.0429778	-0.907104-0.907104
1616	자극정도degree of stimulation	rs55878647rs55878647	PRKCEPRKCE	2:463817572:46381757	A>GA>G	0.0330460.033046	0.04465580.0446558	1.32745011.3274501
1717	자극정도degree of stimulation	rs76118709rs76118709	PRKCEPRKCE	2:463890842:46389084	A>GA>G	0.02155170.0215517	0.04465580.0446558	1.32745011.3274501
1818	자극정도degree of stimulation	rs11125055rs11125055	PRKCEPRKCE	2:463938152:46393815	T>GT>G	0.02442530.0244253	0.04465580.0446558	1.32745011.3274501
1919	자극정도degree of stimulation	rs10041981rs10041981	SPINK5SPINK5	5:1475074205:147507420	T>CT>C	0.42241380.4224138	0.04624190.0462419	-0.332159-0.332159
2020	자극정도degree of stimulation	rs35135425rs35135425	PRKCEPRKCE	2:458793052:45879305	T>CT>C	0.0330460.033046	0.04673010.0467301	1.314931.31493
2121	자극정도degree of stimulation	rs74684800rs74684800	PRKCEPRKCE	2:461091582:46109158	T>CT>C	0.04885060.0488506	0.04859190.0485919	-1.031641-1.031641
2222	자극정도degree of stimulation	rs324420rs324420	FAAHFAAH	1:468707611:46870761	A>CA>C	0.15373560.1537356	0.05170530.0517053	-0.461365-0.461365
2323	자극정도degree of stimulation	rs2295633rs2295633	FAAHFAAH	1:468743831:46874383	A>GA>G	0.15517240.1551724	0.05170530.0517053	0.46136460.4613646
2424	자극정도degree of stimulation	rs17033965rs17033965	PRKCEPRKCE	2:458985302:45898530	A>GA>G	0.23850570.2385057	0.05229350.0522935	0.34407550.3440755
2525	자극정도degree of stimulation	rs12712955rs12712955	PRKCEPRKCE	2:461663212:46166321	A>GA>G	0.17385060.1738506	0.05411160.0541116	-0.434567-0.434567
2626	자극정도degree of stimulation	rs76222090rs76222090	TRPM8TRPM8	2:2349109362:234910936	A>GA>G	0.00287360.0028736	0.05549830.0554983	2.18837262.1883726
2727	자극정도degree of stimulation	rs2030628rs2030628	ABCA12ABCA12	2:2159226122:215922612	T>CT>C	0.21839080.2183908	0.05592890.0559289	-0.428541-0.428541
2828	자극정도degree of stimulation	rs1154469rs1154469	ADH7ADH7	4:1003561794:100356179	T>CT>C	0.2169540.216954	0.05656140.0565614	-0.38814-0.38814
2929	자극정도degree of stimulation	rs11898074rs11898074	PRKCEPRKCE	2:460182232:46018223	A>CA>C	0.44683910.4468391	0.05697810.0569781	0.36122280.3612228
3030	자극정도degree of stimulation	rs16878792rs16878792	NRG1NRG1	8:319508278:31950827	A>GA>G	0.04741380.0474138	0.05748250.0574825	-0.79125-0.79125
3131	자극정도degree of stimulation	rs4952777rs4952777	PRKCEPRKCE	2:460265122:46026512	T>CT>C	0.15086210.1508621	0.07008280.0700828	-0.417561-0.417561
3232	자극정도degree of stimulation	rs6751169rs6751169	TRPM8TRPM8	2:2349113672:234911367	A>GA>G	0.07471260.0747126	0.07209790.0720979	0.65118860.6511886
3333	자극정도degree of stimulation	rs11563054rs11563054	TRPM8TRPM8	2:2349249412:234924941	T>CT>C	0.07471260.0747126	0.07209790.0720979	0.65118860.6511886
3434	자극정도degree of stimulation	rs2711292rs2711292	PRKCEPRKCE	2:461689292:46168929	A>GA>G	0.04022990.0402299	0.0729460.072946	1.03991461.0399146
3535	자극정도degree of stimulation	rs17864759rs17864759	TRPM8TRPM8	2:2349002122:234900212	T>GT>G	0.10775860.1077586	0.07316190.0731619	-0.396413-0.396413
3636	자극정도degree of stimulation	rs28948673rs28948673	TRPM8TRPM8	2:2349014242:234901424	T>CT>C	0.10775860.1077586	0.07316190.0731619	-0.396413-0.396413
3737	자극정도degree of stimulation	rs61763807rs61763807	PRKCEPRKCE	2:462071972:46207197	A>GA>G	0.03160920.0316092	0.07434880.0743488	1.20023711.2002371
3838	자극정도degree of stimulation	rs35043219rs35043219	SPINK5SPINK5	5:1474938335:147493833	T>CT>C	0.42959770.4295977	0.07462270.0746227	0.29492520.2949252
3939	자극정도degree of stimulation	rs6892205rs6892205	SPINK5SPINK5	5:1474753865:147475386	A>GA>G	0.43534480.4353448	0.07561690.0756169	0.29568260.2956826
4040	자극정도degree of stimulation	rs894785rs894785	CTSHCTSH	15:7921486815:79214868	T>GT>G	0.48850570.4885057	0.07842980.0784298	-0.290502-0.290502
4141	자극정도degree of stimulation	rs2888330rs2888330	ABCA12ABCA12	2:2159273292:215927329	A>GA>G	0.25862070.2586207	0.0812890.081289	0.35566150.3556615
4242	자극정도degree of stimulation	rs4953296rs4953296	PRKCEPRKCE	2:462273782:46227378	T>GT>G	0.24137930.2413793	0.0842130.084213	-0.344103-0.344103
4343	자극정도degree of stimulation	rs117238468rs117238468	CTSHCTSH	15:7921769115:79217691	A>GA>G	0.05028740.0502874	0.08544980.0854498	0.68132130.6813213
4444	자극정도degree of stimulation	rs1562653rs1562653	PRKCEPRKCE	2:461969322:46196932	A>CA>C	0.23275860.2327586	0.08633370.0863337	-0.377229-0.377229
4545	자극정도degree of stimulation	rs138402017rs138402017	ABCA12ABCA12	2:2159016882:215901688	A>GA>G	0.00718390.0071839	0.08647940.0864794	1.95476991.9547699
4646	자극정도degree of stimulation	rs10503887rs10503887	NRG1NRG1	8:316334478:31633447	T>GT>G	0.00287360.0028736	0.08647940.0864794	1.95476991.9547699
4747	자극정도degree of stimulation	rs28676960rs28676960	NRG1NRG1	8:316918158:31691815	A>GA>G	0.00287360.0028736	0.08647940.0864794	1.95476991.9547699
4848	자극정도degree of stimulation	rs73232416rs73232416	NRG1NRG1	8:318336518:31833651	T>CT>C	0.00287360.0028736	0.08647940.0864794	1.95476991.9547699
4949	자극정도degree of stimulation	rs78216664rs78216664	NRG1NRG1	8:320683178:32068317	T>GT>G	0.00862070.0086207	0.08647940.0864794	1.95476991.9547699
5050	자극정도degree of stimulation	rs141125665rs141125665	SPINK5SPINK5	5:1474806925:147480692	T>CT>C	0.00718390.0071839	0.08647940.0864794	1.95476991.9547699
5151	자극정도degree of stimulation	rs13008603rs13008603	PRKCEPRKCE	2:463558482:46355848	A>CA>C	0.00718390.0071839	0.08681130.0868113	1.95244641.9524464
5252	자극정도degree of stimulation	rs145574423rs145574423	NRG1NRG1	8:321709688:32170968	T>CT>C	0.03735630.0373563	0.0900380.090038	-0.671861-0.671861
5353	자극정도degree of stimulation	rs10095694rs10095694	NRG1NRG1	8:324997408:32499740	T>CT>C	0.29885060.2988506	0.09262850.0926285	-0.287476-0.287476
5454	자극정도degree of stimulation	rs77790764rs77790764	NRG1NRG1	8:325540008:32554000	A>CA>C	0.03735630.0373563	0.09479830.0947983	-0.564538-0.564538
5555	자극정도degree of stimulation	rs12999695rs12999695	PRKCEPRKCE	2:461604052:46160405	A>GA>G	0.17385060.1738506	0.09564280.0956428	-0.414926-0.414926
5656	자극정도degree of stimulation	rs201328637rs201328637	TGM1TGM1	14:2472897114:24728971	T>CT>C	0.01149430.0114943	0.0973560.097356	1.35981581.3598158

Genetic polymorphism marker predicting presence or absence of hydroquinone stimulation (P < 0.1)

	표현형phenotype	SNP¹⁾ SNP ¹⁾	Gene¹⁾ Gene ¹⁾	Chr:Position¹⁾ Chr:Position ¹⁾	Allele²⁾ Allele ²⁾	MAF³⁾ MAF ³⁾	P-value⁴⁾ P-value ⁴⁾	Effect size (β)⁵⁾ Effect size (β) ⁵⁾
1One	자극유무presence or absence of stimulation	rs11898074rs11898074	PRKCEPRKCE	2:460182232:46018223	A>CA>C	0.44683910.4468391	0.01890.0189	0.8721623360.872162336
22	자극유무presence or absence of stimulation	rs72799991rs72799991	PRKCEPRKCE	2:460186532:46018653	A>CA>C	0.13362070.1336207	0.01980.0198	1.3188435891.318843589
33	자극유무presence or absence of stimulation	rs12999695rs12999695	PRKCEPRKCE	2:461604052:46160405	A>GA>G	0.17385060.1738506	0.02050.0205	-1.13697732-1.13697732
44	자극유무presence or absence of stimulation	rs10166692rs10166692	PRKCEPRKCE	2:460770382:46077038	T>CT>C	0.23994250.2399425	0.02170.0217	0.8119989180.811998918
55	자극유무presence or absence of stimulation	rs6720975rs6720975	PRKCEPRKCE	2:460242122:46024212	T>CT>C	0.2629310.262931	0.02960.0296	-0.751391466-0.751391466
66	자극유무presence or absence of stimulation	rs34597632rs34597632	TGM1TGM1	14:2472806114:24728061	A>GA>G	0.08333330.0833333	0.03220.0322	1.7103436041.710343604
77	자극유무presence or absence of stimulation	rs894785rs894785	CTSHCTSH	15:7921486815:79214868	T>GT>G	0.48850570.4885057	0.03390.0339	-0.66426184-0.66426184
88	자극유무presence or absence of stimulation	rs10095694rs10095694	NRG1NRG1	8:324997408:32499740	T>CT>C	0.29885060.2988506	0.04240.0424	-0.643224983-0.643224983
99	자극유무presence or absence of stimulation	rs11989919rs11989919	NRG1NRG1	8:325026268:32502626	T>CT>C	0.46408050.4640805	0.04250.0425	-0.59529657-0.59529657
1010	자극유무presence or absence of stimulation	rs61588183rs61588183	NRG1NRG1	8:324656318:32465631	T>GT>G	0.47557470.4755747	0.04380.0438	-0.63159702-0.63159702
1111	자극유무presence or absence of stimulation	rs7357509rs7357509	SDR16C5SDR16C5	8:572327878:57232787	T>CT>C	0.04454020.0445402	0.04480.0448	-2.235098679-2.235098679
1212	자극유무presence or absence of stimulation	rs2295633rs2295633	FAAHFAAH	1:468743831:46874383	A>GA>G	0.15517240.1551724	0.05100.0510	0.8711555060.871155506
1313	자극유무presence or absence of stimulation	rs2439279rs2439279	NRG1NRG1	8:324664098:32466409	T>CT>C	0.18103450.1810345	0.05690.0569	-0.814443155-0.814443155
1414	자극유무presence or absence of stimulation	rs4952777rs4952777	PRKCEPRKCE	2:460265122:46026512	T>CT>C	0.15086210.1508621	0.05820.0582	-0.802257509-0.802257509
1515	자극유무presence or absence of stimulation	rs2843018rs2843018	HMGCS2HMGCS2	1:1203110411:120311041	A>GA>G	0.10344830.1034483	0.06080.0608	-0.870864652-0.870864652
1616	자극유무presence or absence of stimulation	rs13401339rs13401339	TRPM8TRPM8	2:2348760372:234876037	A>GA>G	0.33620690.3362069	0.06080.0608	-0.639047644-0.639047644
1717	자극유무presence or absence of stimulation	rs11706136rs11706136	GLB1GLB1	3:330980553:33098055	A>GA>G	0.39655170.3965517	0.06470.0647	-0.578760612-0.578760612
1818	자극유무presence or absence of stimulation	rs13034899rs13034899	PRKCEPRKCE	2:458851192:45885119	T>CT>C	0.48850570.4885057	0.06600.0660	-0.574904223-0.574904223
1919	자극유무presence or absence of stimulation	rs556650rs556650	PRKCEPRKCE	2:459024542:45902454	T>CT>C	0.39511490.3951149	0.06790.0679	0.635188010.63518801
2020	자극유무presence or absence of stimulation	rs75564875rs75564875	CTSHCTSH	15:7923038915:79230389	T>GT>G	0.36494250.3649425	0.06980.0698	-0.537955352-0.537955352
2121	자극유무presence or absence of stimulation	rs117238468rs117238468	CTSHCTSH	15:7921769115:79217691	A>GA>G	0.05028740.0502874	0.07310.0731	1.9426103811.942610381
2222	자극유무presence or absence of stimulation	rs2204204rs2204204	PRKCEPRKCE	2:459059462:45905946	T>CT>C	0.30316090.3031609	0.07720.0772	0.5766223460.576622346
2323	자극유무presence or absence of stimulation	rs7569211rs7569211	TRPM8TRPM8	2:2348821432:234882143	T>CT>C	0.05172410.0517241	0.08010.0801	-1.941060743-1.941060743
2424	자극유무presence or absence of stimulation	rs11080150rs11080150	NF1NF1	17:2962932617:29629326	A>GA>G	0.3620690.362069	0.08050.0805	-0.666008258-0.666008258
2525	자극유무presence or absence of stimulation	rs667226rs667226	HMGCS2HMGCS2	1:1202933521:120293352	T>CT>C	0.16235630.1623563	0.08160.0816	-0.622679889-0.622679889
2626	자극유무presence or absence of stimulation	rs6752001rs6752001	TRPM8TRPM8	2:2349060462:234906046	T>GT>G	0.11350570.1135057	0.08460.0846	-0.769225533-0.769225533
2727	자극유무presence or absence of stimulation	rs7600159rs7600159	PRKCEPRKCE	2:462598072:46259807	A>GA>G	0.26867820.2686782	0.08520.0852	0.6264686850.626468685
2828	자극유무presence or absence of stimulation	rs17033965rs17033965	PRKCEPRKCE	2:458985302:45898530	A>GA>G	0.23850570.2385057	0.08520.0852	0.5872062320.587206232
2929	자극유무presence or absence of stimulation	rs73926053rs73926053	PRKCEPRKCE	2:459775702:45977570	T>CT>C	0.05028740.0502874	0.08930.0893	-1.113100493-1.113100493
3030	자극유무presence or absence of stimulation	rs113407143rs113407143	PRKCEPRKCE	2:462529432:46252943	A>GA>G	0.11637930.1163793	0.09160.0916	-1.122735915-1.122735915
3131	자극유무presence or absence of stimulation	rs4953296rs4953296	PRKCEPRKCE	2:462273782:46227378	T>GT>G	0.24137930.2413793	0.09380.0938	-0.610088516-0.610088516
3232	자극유무presence or absence of stimulation	rs75970876rs75970876	GLB1GLB1	3:330883083:33088308	A>GA>G	0.41235630.4123563	0.09400.0940	0.5115442480.511544248
3333	자극유무presence or absence of stimulation	rs35755034rs35755034	TGM1TGM1	14:2472968714:24729687	T>CT>C	0.06896550.0689655	0.09640.0964	1.3562204411.356220441
3434	자극유무presence or absence of stimulation	rs6982890rs6982890	NRG1NRG1	8:326150988:32615098	T>CT>C	0.17241380.1724138	0.09930.0993	-0.774312808-0.774312808

1) National Institutes of Health (NIH) ID, sequence can be checked on the website

2) (major allele) > (minor allele) meaning

3) minor allele frequency = (2mm + Mm)/2(MM + Mm + mm)

4) Statistical significance of phenotypic differences for the three genotypes (M/M, M/m, m/m) (M: major allele, m: minor allele)

5) As the minor allele increases one by one, the degree of increase or decrease in phenotype* (-: decrease, +: increase)

* Degree of irritation = Inquire about the degree of irritation at the area where hydroquinone was applied using a questionnaire.

* Brightness change = L value among the skin color measurement Lab values of the hydroquinone application area and corresponding area using a colorimeter

Using a commercially available device (chroma meter CR-400, KONIKA MINOLTA, Japan)

실시예 4: 히드로퀴논 효능 및 자극 연관 유의성 유전자 다형성 마커 발굴 및 예측 모델 수립Example 4: Discovery of significant genetic polymorphism markers associated with hydroquinone efficacy and stimulation and establishment of prediction model

4-1: 발굴한 유전자 다형성(SNP) 마커 활용한 히드로퀴논 피부 밝기 예측4-1: Hydroquinone skin brightness prediction using discovered genetic polymorphism (SNP) markers

A hydroquinone skin brightness prediction model was constructed using the discovered markers, and the constructed model can be expressed in the following equation.

Prediction ΔL = Gender variable value + Age variable value +

n: SNP marker combination

beta value: influence factor of the corresponding SNP minor allele

allele type values: Major Major = 1, Major Minor = 2, Minor Minor = 3

Of the 33 people recruited, data obtained from 20 people were used for learning, and data obtained from the remaining 13 people was used to evaluate the performance of the predicted model.

Specifically, the genetic polymorphism markers used in the hydroquinone skin brightness improvement efficacy prediction model are shown in Table 4 below.

Hydroquinone skin brightening improvement efficacy prediction model used genetic polymorphism marker (P < 0.1)

	표현형phenotype	SNP¹⁾ SNP ¹⁾	Gene¹⁾ Gene ¹⁾	Chr:Position¹⁾ Chr:Position ¹⁾	Allele²⁾ Allele ²⁾	MAF³⁾ MAF ³⁾	P-value⁴⁾ P-value ⁴⁾	Effect size (β)⁵⁾ Effect size (β) ⁵⁾
1One	피부 밝기개선Improved skin brightness	rs12788260rs12788260	TYRT.Y.R.	11:8897696311:88976963	T>CT>C	0.04166670.0416667	0.0039315970.003931597	4.0654007954.065400795
22	피부 밝기개선Improved skin brightness	rs4396293rs4396293	TYRT.Y.R.	11:8897611311:88976113	T>CT>C	0.29454020.2945402	6.86E-036.86E-03	-0.991231051-0.991231051
33	피부 밝기개선Improved skin brightness	rs200806138rs200806138	MGST2MGST2	4:1406502724:140650272	A>CA>C	0.02155170.0215517	0.0097677270.009767727	-3.686171454-3.686171454
44	피부 밝기개선Improved skin brightness	rs7162117rs7162117	OCA2OCA2	15:2830434415:28304344	A>GA>G	0.01867820.0186782	0.012305460.01230546	-2.655796929-2.655796929
55	피부 밝기개선Improved skin brightness	rs4855447rs4855447	MITFMITF	3:698033283:69803328	A>GA>G	0.42959770.4295977	0.0191534360.019153436	-1.050123681-1.050123681
66	피부 밝기개선Improved skin brightness	rs2122005rs2122005	OCA2OCA2	15:2826702715:28267027	A>GA>G	0.04166670.0416667	0.0380227290.038022729	1.8318924721.831892472
77	피부 밝기개선Improved skin brightness	rs78955679rs78955679	GSTM5GSTM5	1:1102578141:110257814	C>GC>G	0.1250.125	0.0509516580.050951658	1.5143352391.514335239
88	피부 밝기개선Improved skin brightness	rs2160512rs2160512	MGST1MGST1	12:1651902012:16519020	A>GA>G	0.04741380.0474138	0.0622825690.062282569	-1.208713028-1.208713028
99	피부 밝기개선Improved skin brightness	rs149802322rs149802322	KITLGKITLG	12:8890576612:88905766	A>GA>G	0.01580460.0158046	0.0744242650.074424265	-3.1679249-3.1679249
1010	피부 밝기개선Improved skin brightness	rs202171165rs202171165	GSTM3GSTM3	1:1102829091:110282909	T>CT>C	0.00574710.0057471	8.40E-028.40E-02	2.5430772792.543077279
1111	피부 밝기개선Improved skin brightness	rs4715352rs4715352	GSTA5GSTA5	6:527033966:52703396	T>CT>C	0.21982760.2198276	0.0880029540.088002954	-0.81063853-0.81063853

1) U.S. National Institutes of Health (NIH) ID, sequence can be checked on the website2) (major allele) > (minor allele) Meaning3) minor allele frequency = (2mm + Mm)/2(MM + Mm + mm)

Specifically, R-square, accuracy, sensitivity and specificity were evaluated by comparing actual ΔL and predicted ΔL while increasing the number of markers used in the hydroquinone skin brightness prediction model in Table 4 (Figure 29, Table 5).

Hydroquinone efficacy prediction model (including gender)

마커수Number of markers	마커 조합(n)Marker combination (n)	Adjusted R-square¹⁾ Adjusted R-square ¹⁾	accuracy 정확도²⁾ accuracy Accuracy ²⁾	sensitivity 민감도³⁾ sensitivity Sensitivity ³⁾	specificity 특이도⁴⁾ specificity Specificity ⁴⁾
1One	rs12788260rs12788260	0.29290.2929	0.63640.6364	1.00001.0000	0.00000.0000
22	rs12788260+rs200806138rs12788260+rs200806138	0.44990.4499	0.72730.7273	0.85710.8571	0.50000.5000
33	rs12788260+rs4396293+rs200806138rs12788260+rs4396293+rs200806138	0.55940.5594	0.81820.8182	1.00001.0000	0.50000.5000
......	......	......	......	......	......
66	rs12788260+rs4396293+rs200806138+rs7162117+rs4855447+rs2122005rs12788260+rs4396293+rs200806138+rs7162117+rs4855447+rs2122005	0.68150.6815	0.81820.8182	1.00001.0000	0.50000.5000
......	......	......	......	......	......
1111	rs12788260+rs4396293+rs200806138+rs7162117+rs4855447+rs2122005+rs78955679+rs2160512+rs149802322+rs202171165+rs4715352rs12788260+rs4396293+rs200806138+rs7162117+rs4855447+rs2122005+rs78955679+rs2160512+rs149802322+rs202171165+rs4715352	0.86670.8667	0.90910.9091	1.00001.0000	0.75000.7500

The group with brightness change + is classified as the effective group, and the group with brightness change - is classified as the ineffective group.

1) Adjusted R-square refers to the explanatory power of the prediction model. The closer it is to 1, the better the explanatory power.

2) Accuracy: The degree to which the actual value matches the predicted value.

3) Sensitivity: If the actual value is the efficacy group, the predicted value is the hit predicted as the efficacy group, and if the actual value is +, the predicted value is also the predicted hit as +.

4) Specificity: If the actual value is in the non-efficacy group, the predicted value is also the hit predicted to be in the ineffective group. If the actual value is -, the predicted value is also the hit predicted to be -.

As shown in Table 5 above, it was confirmed that a pattern of increasing accuracy, sensitivity, and specificity appeared as the number of markers used in the hydroquinone skin brightness prediction model increased. This suggests that establishing a prediction model for hydroquinone skin brightening efficacy using multiple relevant markers may help increase the accuracy of predicting hydroquinone skin brightening efficacy.

4-2: 발굴한 유전자 다형성(SNP) 마커 활용한 히드로퀴논 자극 정도 예측4-2: Prediction of degree of hydroquinone stimulation using discovered genetic polymorphism (SNP) markers

A model for predicting the degree of hydroquinone stimulation was constructed using the discovered markers, and the constructed model can be expressed in the following equation.

Predicted stimulus grade = gender variable value + age variable value +

n: SNP marker combination

beta value: influence factor of the corresponding SNP minor allele

allele type values: Major Major = 1, Major Minor = 2, Minor Minor = 3

Of the 93 people recruited, data obtained from 68 people were used for learning, and data obtained from the remaining 25 people was used to evaluate the performance of the predicted model.

Specifically, the genetic polymorphism markers used in the hydroquinone stimulation degree prediction model are shown in Table 6 below.

Genetic polymorphism marker using hydroquinone stimulation degree prediction model (P < 0.1)

	표현형phenotype	SNP¹⁾ SNP ¹⁾	Gene¹⁾ Gene ¹⁾	Chr:Position¹⁾ Chr:Position ¹⁾	Allele²⁾ Allele ²⁾	MAF³⁾ MAF ³⁾	P-value⁴⁾ P-value ⁴⁾	Effect size (β)⁵⁾ Effect size (β) ⁵⁾
1One	자극정도degree of stimulation	rs12712969rs12712969	PRKCEPRKCE	2:463321692:46332169	T>CT>C	0.0301724140.030172414	0.0035010880.003501088	1.374558531.37455853
22	자극정도degree of stimulation	rs34597632rs34597632	TGM1TGM1	14:2472806114:24728061	A>GA>G	0.0833333330.083333333	0.0158130940.015813094	0.7866988930.786698893
33	자극정도degree of stimulation	rs6752001rs6752001	TRPM8TRPM8	2:2349060462:234906046	T>GT>G	0.1135057470.113505747	0.0167807710.016780771	-0.577247735-0.577247735
44	자극정도degree of stimulation	rs3742506rs3742506	TGM1TGM1	14:2472514114:24725141	A>GA>G	0.0747126440.074712644	0.0317113690.031711369	0.6501305740.650130574
55	자극정도degree of stimulation	rs10954810rs10954810	NRG1NRG1	8:315557438:31555743	T>CT>C	0.0071839080.007183908	0.0410771190.041077119	1.371660331.37166033
66	자극정도degree of stimulation	rs11989919rs11989919	NRG1NRG1	8:325026268:32502626	T>CT>C	0.464080460.46408046	0.0428709730.042870973	-0.318473303-0.318473303
77	자극정도degree of stimulation	rs10041981rs10041981	SPINK5SPINK5	5:1475074205:147507420	T>CT>C	0.4224137930.422413793	0.0462418950.046241895	-0.332159359-0.332159359
88	자극정도degree of stimulation	rs74684800rs74684800	PRKCEPRKCE	2:461091582:46109158	T>CT>C	0.0488505750.048850575	0.048591880.04859188	-1.031640874-1.031640874
99	자극정도degree of stimulation	rs17033965rs17033965	PRKCEPRKCE	2:458985302:45898530	A>GA>G	0.2385057470.238505747	0.0522935430.052293543	0.344075520.34407552
1010	자극정도degree of stimulation	rs12712955rs12712955	PRKCEPRKCE	2:461663212:46166321	A>GA>G	0.1738505750.173850575	0.054111650.05411165	-0.434566768-0.434566768
1111	자극정도degree of stimulation	rs76222090rs76222090	TRPM8TRPM8	2:2349109362:234910936	A>GA>G	0.0028735630.002873563	0.0554983420.055498342	2.1883725932.188372593
1212	자극정도degree of stimulation	rs11898074rs11898074	PRKCEPRKCE	2:460182232:46018223	A>CA>C	0.446839080.44683908	0.0569781450.056978145	0.3612227580.361222758
1313	자극정도degree of stimulation	rs16878792rs16878792	NRG1NRG1	8:319508278:31950827	A>GA>G	0.0474137930.047413793	0.0574825470.057482547	-0.791249875-0.791249875
1414	자극정도degree of stimulation	rs2711292rs2711292	PRKCEPRKCE	2:461689292:46168929	A>GA>G	0.0402298850.040229885	0.0729459540.072945954	1.0399145931.039914593
1515	자극정도degree of stimulation	rs1562653rs1562653	PRKCEPRKCE	2:461969322:46196932	A>CA>C	0.2327586210.232758621	0.0863336940.086333694	-0.377229356-0.377229356
1616	자극정도degree of stimulation	rs145574423rs145574423	NRG1NRG1	8:321709688:32170968	T>CT>C	0.0373563220.037356322	0.0900379540.090037954	-0.671861336-0.671861336
1717	자극정도degree of stimulation	rs10095694rs10095694	NRG1NRG1	8:324997408:32499740	T>CT>C	0.2988505750.298850575	0.0926284520.092628452	-0.287476276-0.287476276
1818	자극정도degree of stimulation	rs201328637rs201328637	TGM1TGM1	14:2472897114:24728971	T>CT>C	0.0114942530.011494253	0.0973560010.097356001	1.3598157851.359815785

1) U.S. National Institutes of Health (NIH) ID, sequence can be checked on the website 2) (major allele) > (minor allele) Meaning

3) minor allele frequency = (2mm + Mm)/2(MM + Mm + mm)

In addition, the classification criteria and number of irritation grades of hydroquinone are shown in Table 7 below.

GradeGrade	자극 유무presence or absence of stimulation	자극 Grade 기준 Based on stimulation grade		인원personnel
00	자극 無No irritation	자극 전혀 없음No irritation at all	4040
1One	자극 有There is stimulation	미미 (초기 5일내, 또는 평가기간 내 1~2회 간헐적 자극)Minor (intermittent stimulation within the first 5 days or 1-2 times during the evaluation period)	2222
22		약함 (7일 이상 자극 지속되나 견디고 지속 사용)Mild (irritation persists for more than 7 days, but tolerate and continue use)	1515
33		강함 (자극으로 평가 중단 및 피험자가 자극을 매우 강조)Strong (stimulus interrupts evaluation and subject places great emphasis on stimulus)	1616

Specifically, R-square, accuracy, sensitivity, and specificity were evaluated by comparing the actual stimulation grade and the predicted stimulation grade while increasing the number of markers used in the hydroquinone stimulation degree prediction model in Table 6 (Figure 30, Table 8 ).

Hydroquinone irritation level prediction model (including gender)

마커수Number of markers	마커 조합 (n)Marker combination (n)	Adjusted R-square¹⁾ Adjusted R-square ¹⁾	accuracy 정확도²⁾ accuracy Accuracy ²⁾	sensitivity 민감도³⁾ sensitivity Sensitivity ³⁾	specificity 특이도⁴⁾ specificity Specificity ⁴⁾
1One	rs12712969rs12712969	0.09310.0931	0.68000.6800	0.22220.2222	0.93750.9375
......	......	......	......	......	......
55	rs12712969+rs34597632+rs6752001+rs3742506+rs10954810rs12712969+rs34597632+rs6752001+rs3742506+rs10954810	0.25560.2556	0.72000.7200	0.44440.4444	0.87500.8750
......	......	......	......	......	......
1010	rs12712969+rs34597632+rs6752001+rs3742506+rs10954810+rs11989919+rs10041981+rs74684800+rs17033965+rs12712955rs12712969+rs34597632+rs6752001+rs3742506+rs10954810+rs11989919+rs10041981+rs74684800+rs17033965+rs12712955	0.35030.3503	0.76000.7600	0.44440.4444	0.93750.9375
......	......	......	......	......	......
1818	rs12712969+rs34597632+rs6752001+rs3742506+rs10954810+rs11989919+rs10041981+rs74684800+rs17033965+rs12712955+rs76222090+rs11898074+rs16878792+rs2711292+rs1562653+rs145574423+rs10095694+rs201328637rs12712969+rs34597632+rs6752001+rs3742506+rs10954810+rs11989919+rs10041981+rs74684800+rs17033965+rs12712955+rs76222090+rs11898074+rs16 878792+rs2711292+rs1562653+rs145574423+rs10095694+rs201328637	0.53610.5361	0.84000.8400	0.55560.5556	1.00001.0000

The group with stimulation grade 2~3 is classified as stimulation group, and the group with stimulation grade 0~1 is classified as non-stimulation group.

1) Adjusted R-square refers to the explanatory power of the prediction model, and the closer it is to 1, the better the explanatory power.

2) Accuracy: The degree to which the actual value matches the predicted value.

3) Sensitivity: If the actual value is the stimulus group, the predicted value is also the hit predicted by the stimulus group. If the actual value is +, the predicted value is also the hit predicted as +.

4) Specificity: If the actual value is the non-stimulation group, the predicted value is also the hit predicted as the non-stimulation group. If the actual value is -, the predicted value is also the hit predicted as -.

As shown in Table 8 above, it was confirmed that accuracy, sensitivity, and specificity increased as the number of markers used in the hydroquinone stimulation degree prediction model increased. This suggests that establishing a hydroquinone stimulation prediction model using multiple relevant markers can help increase the accuracy of predicting the degree of hydroquinone stimulation.

4-3: 발굴한 유전자 다형성(SNP) 마커 활용한 히드로퀴논 자극 유무 예측4-3: Prediction of hydroquinone stimulation using discovered genetic polymorphism (SNP) markers

A prediction model for the presence or absence of hydroquinone stimulation was constructed using the discovered markers, and the constructed model can be expressed in the following equation.

Presence or absence of predicted stimulus = gender variable value + age variable value +

n: SNP marker combination

beta value: influence factor of the corresponding SNP minor allele

allele type values: Major Major = 1, Major Minor = 2, Minor Minor = 3

Specifically, the genetic polymorphism markers used in the prediction model for the presence or absence of hydroquinone stimulation are shown in Table 9 below.

Genetic polymorphism marker using hydroquinone stimulation prediction model (P < 0.1)

	표현형phenotype	SNP¹⁾ SNP ¹⁾	Gene¹⁾ Gene ¹⁾	Chr:Position¹⁾ Chr:Position ¹⁾	Allele²⁾ Allele ²⁾	MAF³⁾ MAF ³⁾	P-value⁴⁾ P-value ⁴⁾	Effect size (β)⁵⁾ Effect size (β) ⁵⁾
1One	자극유무presence or absence of stimulation	rs10166692rs10166692	PRKCEPRKCE	2:460770382:46077038	T>CT>C	0.23994250.2399425	0.02170.0217	0.8119989180.811998918
22	자극유무presence or absence of stimulation	rs6720975rs6720975	PRKCEPRKCE	2:460242122:46024212	T>CT>C	0.2629310.262931	0.02960.0296	-0.751391466-0.751391466
33	자극유무presence or absence of stimulation	rs34597632rs34597632	TGM1TGM1	14:2472806114:24728061	A>GA>G	0.08333330.0833333	0.03220.0322	1.7103436041.710343604
44	자극유무presence or absence of stimulation	rs2295633rs2295633	FAAHFAAH	1:468743831:46874383	A>GA>G	0.15517240.1551724	0.05100.0510	0.8711555060.871155506
55	자극유무presence or absence of stimulation	rs13401339rs13401339	TRPM8TRPM8	2:2348760372:234876037	A>GA>G	0.33620690.3362069	0.06080.0608	-0.639047644-0.639047644
66	자극유무presence or absence of stimulation	rs11706136rs11706136	GLB1GLB1	3:330980553:33098055	A>GA>G	0.39655170.3965517	0.06470.0647	-0.578760612-0.578760612
77	자극유무presence or absence of stimulation	rs117238468rs117238468	CTSHCTSH	15:7921769115:79217691	A>GA>G	0.05028740.0502874	0.07310.0731	1.9426103811.942610381
88	자극유무presence or absence of stimulation	rs7569211rs7569211	TRPM8TRPM8	2:2348821432:234882143	T>CT>C	0.05172410.0517241	0.08010.0801	-1.941060743-1.941060743
99	자극유무presence or absence of stimulation	rs6752001rs6752001	TRPM8TRPM8	2:2349060462:234906046	T>GT>G	0.11350570.1135057	0.08460.0846	-0.769225533-0.769225533
1010	자극유무presence or absence of stimulation	rs113407143rs113407143	PRKCEPRKCE	2:462529432:46252943	A>GA>G	0.11637930.1163793	0.09160.0916	-1.122735915-1.122735915
1111	자극유무presence or absence of stimulation	rs4953296rs4953296	PRKCEPRKCE	2:462273782:46227378	T>GT>G	0.24137930.2413793	0.09380.0938	-0.610088516-0.610088516
1212	자극유무presence or absence of stimulation	rs75970876rs75970876	GLB1GLB1	3:330883083:33088308	A>GA>G	0.41235630.4123563	0.09400.0940	0.5115442480.511544248
1313	자극유무presence or absence of stimulation	rs6982890rs6982890	NRG1NRG1	8:326150988:32615098	T>CT>C	0.17241380.1724138	0.09930.0993	-0.774312808-0.774312808

1) National Institutes of Health (NIH) ID, sequence can be checked on the website2) (major allele) > (minor allele) Meaning3) minor allele frequency = (2mm + Mm)/2(MM + Mm + mm)

Specifically, while increasing the number of markers used in the hydroquinone stimulation prediction model in Table 9, R-square, accuracy, sensitivity, and specificity were evaluated by comparing the actual stimulation with the predicted stimulation (Figure 31, Table 10) ).

When applying a model to predict the presence or absence of hydroquinone stimulation through marker combination (data processed as no stimulation: 0, stimulation: 1)

Hydroquinone stimulation presence/absence prediction model (including gender)

마커수Number of markers	마커 조합(n)Marker combination (n)	Adjusted R-square¹⁾ Adjusted R-square ¹⁾	accuracy 정확도²⁾ accuracy Accuracy ²⁾	sensitivity 민감도³⁾ sensitivity Sensitivity ³⁾	specificity 특이도⁴⁾ specificity Specificity ⁴⁾
1One	rs10166692rs10166692	0.00640.0064	1One	1One	1One
……	……	……	……	……	……
55	rs10166692+rs6720975+rs34597632+rs2295633+rs13401339rs10166692+rs6720975+rs34597632+rs2295633+rs13401339	0.15280.1528	0.82000.8200	0.87500.8750	1One
……	……	……	……	……	……
77	rs10166692+rs6720975+rs34597632+rs2295633+rs13401339+rs11706136+rs117238468rs10166692+rs6720975+rs34597632+rs2295633+rs13401339+rs11706136+rs117238468	0.22660.2266	0.88000.8800	0.81250.8125	1One
……	……	……	……	……	……
1313	rs10166692+rs6720975+rs34597632+rs2295633+rs13401339+rs11706136+rs117238468+rs7569211+rs6752001+rs113407143+rs4953296+rs75970876+rs6982890rs10166692+rs6720975+rs34597632+rs2295633+rs13401339+rs11706136+rs117238468+rs7569211+rs6752001+rs113407143+rs4953296+rs75970876+rs698 2890	0.44210.4421	0.84000.8400	0.75000.7500	1One

The group with a predicted value of 0.5 or more is classified as a stimulus group, and the group with a predicted value of less than 0.5 is classified as a non-stimulus group.1) Adjusted R-square means the explanatory power of the prediction model, and the closer it is to 1, the better the explanatory power.2) Accuracy: The actual value and Degree to which predictions agree

3) Sensitivity: If the actual value is the stimulus group, the degree to which the predicted value is close to the stimulus group.

4) Specificity: If the actual value is in the non-stimulation group, the predicted value is close to the non-stimulation group.

As shown in Table 10 above, it was confirmed that a pattern of increasing accuracy, sensitivity, and specificity appeared as the number of markers used in the model for predicting the presence or absence of hydroquinone stimulation increased. This suggests that establishing a hydroquinone stimulation prediction model using multiple relevant markers can help increase the accuracy of predicting the presence or absence of hydroquinone stimulation.

실시예 5: 발굴한 유전자 다형성 (SNP) 마커에 따른 히드로퀴논 자극 완화 효과 확인Example 5: Confirmation of hydroquinone irritation relief effect according to discovered genetic polymorphism (SNP) marker

A total of 7 out of 18 SNPs (Table 6) used in the stimulation prediction model created through SNP markers derived from correlation analysis between the irritation grade (0-3) caused by the hydroquinone 4% product in 93 subjects and genes related to the hydroquinone stimulation mechanism. The gene (PKRCE) containing the SNP of the species was selected.

Prediction of irritation by PRKCE using 7 SNP types of PRKCE gene We sought to discover substances that inhibit irritation of PRKCE according to grade and prescribe customized irritation relievers (Table 11).

Specifically, Western red pepper herb extract and Salvia Salvia root extract were used as irritation relievers.

Genes (SNPs) used to verify the efficacy of hydroquinone in reducing irritation and ingredients that inhibit the stimulation mechanism

GeneGene	유전자 자극 메커니즘gene stimulation mechanism	SNPSNP	AlleleAllele	MAFMAF	P-valueP-value	Effect size (β)Effect size (β)	자극 발생 억제 성분Ingredients that inhibit irritation
PRKCEPRKCE	-캡사이신 인지 수용체인 TRPV1이 PRKCE에 의한 phosphorylation을 통해 활성화되어 통증 유발-TRPV1, a capsaicin recognition receptor, is activated through phosphorylation by PRKCE, causing pain.	rs17033965rs17033965	A>GA>G	0.23850.2385	5.22E-025.22E-02	0.34410.3441	-서양고추나물추출물 -단삼뿌리추출물-Western red pepper herb extract -Salvia Salvia root extract
		rs11898074rs11898074	A>CA>C	0.44680.4468	5.70E-025.70E-02	0.36120.3612
		rs74684800rs74684800	T>CT>C	0.04890.0489	4.86E-024.86E-02	-1.0316-1.0316
		rs12712955rs12712955	A>GA>G	0.17390.1739	5.41E-025.41E-02	-0.4346-0.4346
		rs2711292rs2711292	A>GA>G	0.04020.0402	7.29E-027.29E-02	1.03991.0399
		rs1562653rs1562653	A>CA>C	0.23270.2327	8.63E-028.63E-02	-0.3772-0.3772
		rs12712969rs12712969	T>CT>C	0.03010.0301	3.50E-023.50E-02	1.37451.3745

5-1. PRKCE 억제 물질을 포함한 히드로퀴논 자극 완화 겔 제조5-1. Preparation of hydroquinone irritation relief gel containing PRKCE inhibitory substance

An irritation relief gel was prepared containing 1.5% by weight of Western red pepper herb extract and 0.075% by weight of Salvia Salvia root extract, which are PRKCE inhibitors.

5-2. PRKCE SNP 기반 예측 자극 grade에 따른 자극 완화겔의 자극 완화 효과 차이5-2. Differences in irritation relief effect of irritation relief gel according to PRKCE SNP-based predicted stimulation grade

When applying a hydroquinone 4% product, 15 subjects corresponding to the irritation grade (1-3) were asked to apply the irritation relief gel to the local area of the face once a day (in the evening) for 3 weeks and allow it to be absorbed, followed by one type of hydroquinone 4% product. was applied to local areas of the face once a day (evening). Afterwards, a questionnaire was conducted regarding irritation including itchiness, stinging, erythema, and rash, and evaluation was conducted on a 4-point scale (0: no irritation at all, 1: slight irritation, 2: weak, 3: strong) to determine whether the hydroquinone 4% product was used alone. The irritation relieving effect was confirmed by checking whether the grade decreased when applying the irritation relieving gel complex compared to the irritation grade at the time of application.

(Irritation relieving effect = irritation grade when applying 4% hydroquinone product alone - irritation grade when applying 4% hydroquinone product + irritation relieving gel combined) (However, if the value is negative, it is replaced with 0)

Specifically, the PRKCE SNP-based stimulation prediction grade of 15 subjects was calculated and the PRKCE stimulation high-risk/low-risk group was divided into high-risk and low-risk groups based on the average value of 1.534 to confirm the difference in stimulation relief by group. PRKCE inhibition according to the predicted stimulation grade based on the seven types of PRKCE SNPs in Table 9 As a result of comparing the irritation relief effects of irritation relief gels containing PRKCE SNP substances, it was confirmed that the higher the predicted stimulation grade based on PRKCE SNP, the better the stimulation relief effect of irritation relief gels containing PRKCE inhibitory substances (Table 12, Figure 32).

Changes in irritation grade due to combined application of irritation relieving gel

피험자　subject	자극 예측 gradeStimulus prediction grade	PRKCE에 의한 자극 예측Stimulus prediction by PRKCE	히드로퀴논 4% 제품 단독 도포 자극 gradeHydroquinone 4% product applied alone is irritating grade	히드로퀴논 4% 제품+ 자극 완화겔 복합 도포 자극 gradeHydroquinone 4% product + irritation relief gel combined application irritation grade	PRKCE 억제 소재 자극완화겔의 자극 완화 효과Irritation-relieving effect of PRKCE-inhibiting material irritation-relieving gel	PRKCE 억제 소재 자극완화겔의 자극 감소 인지 여부Whether or not the stimulation of the irritation-relieving gel, which is a material that inhibits PRKCE, is reduced
LG001LG001	2.862.86	고위험 high risk	22	22	00	XX
LG002LG002	2.652.65		33	1One	22	감소decrease
LG003LG003	2.502.50		1One	00	1One	감소decrease
LG004LG004	2.382.38		33	22	1One	감소decrease
LG005LG005	1.781.78		1One	00	1One	감소decrease
LG006LG006	1.651.65		33	22	1One	감소decrease
LG007LG007	1.641.64		22	00	22	감소decrease
LG008LG008	1.431.43	저위험low risk	1One	1One	00	XX
LG009LG009	1.001.00		1One	00	1One	감소decrease
LG010LG010	1.001.00		1One	22	00	XX
LG011LG011	0.960.96		1One	1One	00	XX
LG012LG012	0.930.93		1One	1One	00	XX
LG013LG013	0.870.87		1One	00	1One	감소decrease
LG014LG014	0.680.68		1One	1One	00	XX
LG015LG015	0.680.68		1One	22	00	XX

Through this, it can be seen that the hydroquinone stimulation-alleviating effect of the stimulus-related gene function inhibitor component selectively varies depending on the stimulus prediction gene type.

From the above description, those skilled in the art to which the present invention pertains will understand that the present invention can be implemented in other specific forms without changing its technical idea or essential features. In this regard, the embodiments described above should be understood in all respects as illustrative and not restrictive. The scope of the present invention should be construed as including the meaning and scope of the patent claims described below rather than the detailed description above, and all changes or modified forms derived from the equivalent concept thereof are included in the scope of the present invention.

Claims

(a) inputting SNP information of the prediction target entity into the prediction terminal; and

(b) the prediction terminal queries the SNP information to a skin brightness change prediction model pre-stored in the prediction terminal or server to calculate predicted skin brightness change information, and outputs the calculated predicted skin brightness change information; Contains,

The skin brightness change prediction model is,

The learning processor learns first skin brightness change learning data consisting of a pair of SNP information and skin brightness change information when a substance containing hydroquinone is applied to an object having the SNP information, and when SNP information is queried, the queried SNP A first skin brightness change prediction model that outputs first predicted skin brightness change information when a substance containing hydroquinone is applied to an object having the information,

Method for predicting changes in skin brightness of prescriptions containing hydroquinone.
According to paragraph 1,

The SNPs marker selection device further includes selecting one or more SNPs associated with a change in skin brightness upon application of hydroquinone from among a plurality of SNPs included in the SNP information,

The SNP information of the learning data includes at least some of the one or more SNPs selected by the SNPs marker selection device, and does not include SNPs not selected by the SNPs marker selection device.

Method for predicting changes in skin brightness of prescriptions containing hydroquinone.
According to paragraph 2,

One or more SNPs related to the change in skin brightness are one or more selected from Table 1,

Method for predicting changes in skin brightness of prescriptions containing hydroquinone.
According to paragraph 1,

As the learning process learns the learning data, a weight that affects the change in skin brightness when applying a substance containing hydroquinone can be determined for each SNP included in the SNP information,

The weight is,

A first value determined according to the number of major alleles and minor alleles included in each SNP; and

It is calculated using a second value whose value changes through learning,

Method for predicting changes in skin brightness of prescriptions containing hydroquinone.
According to clause 4,

The first value is,

Major allele - major allele, major allele - minor allele, and minor allele - different values are assigned to each SNP type of the minor allele,

Method for predicting changes in skin brightness of prescriptions containing hydroquinone.
According to clause 5,

The first value has a higher value for types with a greater number of lower alleles among the SNPs types.

Method for predicting changes in skin brightness of prescriptions containing hydroquinone.
According to clause 4,

The first value is a value that does not change even if learning occurs,

Method for predicting changes in skin brightness of prescriptions containing hydroquinone.
According to paragraph 4,

The weight is a product of the first value and the second value,

The predicted skin brightness change information is calculated by summing the weights for each SNP,

Method for predicting changes in skin brightness of prescriptions containing hydroquinone.
According to paragraph 1,

The skin brightness change prediction model is,

The learning processor consists of a pair of SNP information, one or more prescription information among a plurality of prescriptions including a prescription containing hydroquinone, and skin brightness change information when a substance according to the prescription is applied to an individual having the SNP information. 2 A second skin brightness that learns skin brightness change learning data and, when SNP information is queried, outputs second predicted skin brightness change information when each substance according to a plurality of prescriptions is applied to an individual with the queried SNP information. A change prediction model,

Method for predicting changes in skin brightness of prescriptions containing hydroquinone.
According to paragraph 1,

The skin brightness change prediction model is,

In addition to SNP information, the learning processor provides one or more of age information, gender information, lifestyle information, environmental information, and skin characteristic information, and information on changes in skin brightness when a substance containing hydroquinone is applied to an individual having the SNP information. By learning the third skin brightness change learning data consisting of pairs, when one or more of SNP information, age information, gender information, lifestyle information, environmental information, and skin characteristic information is queried, hydroquinone is applied to the object having the queried information. A third skin brightness change prediction model that outputs third predicted skin brightness change information when a material containing is applied,

Method for predicting changes in skin brightness of prescriptions containing hydroquinone.
According to paragraph 1,

The skin brightness change prediction model is,

In addition to the SNP information, the learning processor may include one or more of age information, gender information, lifestyle information, environmental information, and skin characteristic information, one or more prescription information of a plurality of prescriptions including a prescription containing hydroquinone, and the SNP information. By learning the fourth skin brightness change learning data consisting of pairs of skin brightness change information when the substance according to the prescription is applied to an object having, SNP information, age information, gender information, lifestyle information, environmental information, and skin When one or more pieces of characteristic information are queried, a fourth skin brightness change prediction model that outputs fourth predicted skin brightness change information when each substance according to a plurality of prescriptions is applied to an object having the queried information,

Method for predicting changes in skin brightness of prescriptions containing hydroquinone.
According to claim 9 or 11,

In step (b),

(b) the prediction terminal queries the skin brightness change prediction model for the SNP information to calculate predicted skin brightness change information for each of a plurality of prescriptions;

(c) further comprising the step of the prescription selection unit selecting one or more prescriptions among a plurality of prescriptions using the predicted skin brightness change information,

Method for predicting changes in skin brightness of prescriptions containing hydroquinone.
According to clause 12,

In step (c),

Further comprising the step of the prescription selection unit 33 selecting one or more prescriptions with a large value of the predicted skin brightness change information from among the plurality of prescriptions,

Method for predicting changes in skin brightness of prescriptions containing hydroquinone.
According to claim 9 or 11,

The plurality of prescriptions include hydroquinone prescriptions and prescriptions containing additional substances to hydroquinone,

The additional substance is one or more selected from the group consisting of a skin brightening increase inducer, a skin irritation reliever, a substance stabilizing substance, and a substance texture determining substance.

Method for predicting changes in skin brightness of prescriptions containing hydroquinone.
According to paragraph 1,

In step (b),

The prediction terminal queries the SNP information to a skin irritation prediction model pre-stored in the prediction terminal or the server, and the calculated information is additionally output, or is determined in a threshold irritation score prediction model pre-stored in the prediction terminal or the server. Further comprising the step of additionally outputting a threshold stimulus score,

In the skin irritation prediction model, the learning processor learns first stimulus learning data consisting of a pair of SNP information and stimulus information when a substance containing hydroquinone is applied to an object having the SNP information, and the SNP information is used as a query. If so, it is a model that calculates the first predicted stimulus information when a substance containing hydroquinone is applied to an individual with the queried SNP information,

The critical stimulation score prediction model is formed by the learning processor learning first critical stimulation learning data consisting of a stimulation score predetermined according to SNP information and a pair of stimulation presence or absence when hydroquinone is applied to an entity having the SNP information, Determining the threshold irritation score that causes skin irritation when a substance containing hydroquinone is applied,

Method for predicting changes in skin brightness of prescriptions containing hydroquinone.
According to clause 15,

According to the comparison result between the output critical stimulation score and the stimulation score determined according to the queried SNP information, the second predicted stimulation information when a substance containing hydroquinone is applied to an individual having the queried SNP information is calculated. Including further steps,

Method for predicting changes in skin brightness of prescriptions containing hydroquinone.
According to clause 16,

(c) An individual having queried SNP information among a plurality of prescriptions containing hydroquinone, where the prescription selection unit uses one or more of the predicted skin brightness change information, the first predicted stimulus information, and the second predicted stimulus information. Further comprising: selecting one or more prescriptions suitable to be prescribed,

Method for predicting changes in skin brightness of prescriptions containing hydroquinone.
A server that generates a first skin brightness change prediction model by learning first skin brightness change learning data consisting of a pair of SNP information and skin brightness change information when a substance containing hydroquinone is applied to an individual having the SNP information, When SNP information is queried, a server that generates a first skin brightness change prediction model that outputs first predicted skin brightness change information when a substance containing hydroquinone is applied to an individual having the queried SNP information; and

A prediction terminal that receives SNP information of a prediction target entity and outputs first predicted skin brightness change information calculated by querying the SNP information input to the first skin brightness change prediction model generated by the server.

A system for predicting changes in skin brightness of prescriptions containing hydroquinone.
According to clause 18,

It further includes a SNPs marker selection device for selecting one or more SNPs related to changes in skin brightness upon application of hydroquinone among the plurality of SNPs included in the SNP information,

The SNP information of the learning data includes at least some of the one or more SNPs selected by the SNPs marker selection device, and does not include SNPs not selected by the SNPs marker selection device.

A system for predicting changes in skin brightness of prescriptions containing hydroquinone.
According to clause 19,

One or more SNPs related to the change in skin brightness are one or more selected from Table 1,

A system for predicting changes in skin brightness of prescriptions containing hydroquinone.
According to clause 18,

As the server's learning process learns the learning data, a weight that affects the change in skin brightness when applying a substance containing hydroquinone can be determined for each SNP included in the SNP information,

The weight is,

A first value determined according to the number of major alleles and minor alleles included in each SNP; and

It is calculated using a second value whose value changes through learning,

A system for predicting changes in skin brightness of prescriptions containing hydroquinone.
According to clause 21,

The first value is,

Major allele - major allele, major allele - minor allele, and minor allele - different values are assigned to each SNP type of the minor allele,

A system for predicting changes in skin brightness of prescriptions containing hydroquinone.
According to clause 22,

The first value has a higher value for types with a greater number of lower alleles among the SNPs types.

A system for predicting changes in skin brightness of prescriptions containing hydroquinone.
According to clause 21,

The first value is a value that does not change even if learning occurs,

A system for predicting changes in skin brightness of prescriptions containing hydroquinone.
According to clause 21,

The weight is a product of the first value and the second value,

The predicted skin brightness change information is calculated by summing the weights for each SNP,

A system for predicting changes in skin brightness of prescriptions containing hydroquinone.
According to clause 18,

The skin brightness change prediction model is,

The server is a pair of SNP information, one or more prescription information among a plurality of prescriptions including a prescription containing hydroquinone, and skin brightness change information when a substance according to the prescription is applied to an individual having the SNP information. A second pair consisting of A second skin brightness change that learns skin brightness change learning data and, when SNP information is queried, outputs second predicted skin brightness change information when each substance according to a plurality of prescriptions is applied to an individual with the queried SNP information. As a prediction model,

A system for predicting changes in skin brightness of prescriptions containing hydroquinone.
According to clause 18,

The skin brightness change prediction model is,

In addition to the SNP information, the server provides one or more of age information, gender information, lifestyle information, environmental information, and skin characteristic information, and information on changes in skin brightness when a substance containing hydroquinone is applied to an individual having the SNP information. By learning the third skin brightness change learning data consisting of pairs, when one or more of SNP information, age information, gender information, lifestyle information, environmental information, and skin characteristic information is queried, hydroquinone is applied to the object having the queried information. A third skin brightness change prediction model that outputs third predicted skin brightness change information when a material containing is applied,

A system for predicting changes in skin brightness of prescriptions containing hydroquinone.
According to clause 18,

The skin brightness change prediction model is,

In addition to the SNP information, the learning processor may include one or more of age information, gender information, lifestyle information, environmental information, and skin characteristic information, one or more prescription information of a plurality of prescriptions including a prescription containing hydroquinone, and the SNP information. By learning the fourth skin brightness change learning data consisting of pairs of skin brightness change information when the substance according to the prescription is applied to an object having, SNP information, age information, gender information, lifestyle information, environmental information, and skin A fourth skin brightness change prediction model that outputs fourth predicted skin brightness change information when each of the substances according to a plurality of prescriptions is applied to an object having the queried information when one or more pieces of characteristic information are queried,

A system for predicting changes in skin brightness of prescriptions containing hydroquinone.
According to claim 26 or 28,

The prediction terminal queries the skin brightness change prediction model for the SNP information and calculates predicted skin brightness change information for each of a plurality of prescriptions,

The system is,

Further comprising a prescription selection unit that selects one or more prescriptions from a plurality of prescriptions using the predicted skin brightness change information,

A system for predicting changes in skin brightness of prescriptions containing hydroquinone.
According to clause 29,

The prescription selection unit selects one or more prescriptions with a large value of the predicted skin brightness change information from among the plurality of prescriptions,

A system for predicting changes in skin brightness of prescriptions containing hydroquinone.
According to claim 26 or 28,

The plurality of prescriptions include hydroquinone prescriptions and prescriptions containing additional substances to hydroquinone,

The additional substance is one or more selected from the group consisting of a skin brightening increase inducer, a skin irritation reliever, a substance stabilizing substance, and a substance texture determining substance.

A system for predicting changes in skin brightness of prescriptions containing hydroquinone.
According to clause 18,

The prediction terminal additionally outputs information calculated by querying the SNP information input to the skin irritation prediction model generated by the server, or additionally outputs a score determined in the threshold irritation score prediction model generated by the server,

The skin irritation prediction model learns first stimulus learning data consisting of pairs of stimulus information when the learning processor applies SNP information and a substance containing hydroquinone to an object having the SNP information, and the SNP information is queried. In this case, it is a model that calculates the first predicted stimulus information when a substance containing hydroquinone is applied to an individual with the queried SNP information,

The critical stimulation score prediction model is configured by the learning processor to learn first critical stimulation learning data consisting of a stimulation score predetermined according to SNP information and a pair of stimulation presence or absence when hydroquinone is applied to an individual having the SNP information, Outputting a threshold irritation score that causes skin irritation when a substance containing hydroquinone is applied,

A system for predicting changes in skin brightness of prescriptions containing hydroquinone.
According to clause 32,

The prediction terminal makes a second prediction when a substance containing hydroquinone is applied to an individual having the queried SNP information according to a comparison result between the output critical stimulus score and the stimulus score determined according to the queried SNP information. Further calculating stimulus information,

A system for predicting changes in skin brightness of prescriptions containing hydroquinone.
According to clause 33,

One or more prescriptions suitable to be prescribed to an individual having the queried SNP information among a plurality of prescriptions containing hydroquinone using the skin brightness change information, the first predictive stimulus information, and the second predictive stimulus information. Further comprising a prescription selection unit for selecting,

A system for predicting changes in skin brightness of prescriptions containing hydroquinone.
A computer program stored in a computer-readable recording medium for executing the method according to any one of claims 1 to 11 and 15 to 17.